Clinical Inquiries

Evidence summary

Surgical repair: Re-injury risk goes down, complications risk goes up

A 2021 network meta-analysis including 38 RCTs (N = 2480) reported outcomes in patients ages 18 and older with acute Achilles tendon rupture (AATR) and 3 or more months of follow-up.¹ A significant increase in re-rupture rate was shown in patients who underwent nonoperative vs open repair (risk ratio [RR] = 2.41; 95% CI, 1.12-5.18). There was a significant decrease in wound-related complications in nonoperative vs open-repair patients (RR = 0.23; 95% CI, 0.06-0.88). There was also a significant difference in incidence of sural nerve injury in nonoperative vs open repair (RR = 0.27; 95% CI, 0.08-0.94). There were no significant differences in return to sport between open repair and nonoperative repair (RR = 0.62; 95% CI, 0.22-1.77). Insufficient data were reported to calculate the number needed to treat (NNT) and number needed to harm (NNH) for these outcomes.

Additionally, the authors looked at traditional standard rehabilitation and accelerated functional rehabilitation in both the operative and the nonoperative setting. The type of rehabilitation program did not have a significant impact on complications of re-rupture, wound, or sural nerve injury.

The included studies had an overall low risk of publication bias based on Begg’s funnel plot test (Pr > |z| = 0.86). The highest risk was performance bias, as neither the participants nor personnel were blinded to treatment in 71% of the studies.

Functional outcomes are similar for surgical vs nonoperative repair

In a 2019 meta-analysis of 9 RCTs (N = 822), adults ages 18 and older with AATR and a minimum of 12 months’ follow-up were randomized to either operative or nonoperative repair. There was a decreased rate of rupture with surgical repair and an associated increased rate of complications (ie, superficial wound infections and nerve injury). However, there was no significant difference in Physical Activity Scale (PAS) score between the 2 groups (mean difference, –0.05; 95% CI, –0.37 to 0.27).² With surgical intervention, the NNT for Achilles tendon re-rupture was 15, and the NNH for superficial wound infection and nerve injury, respectively, were 22 and 28. Limitations of the study included different operative techniques and rehab protocols, which may have affected the results of the included studies.

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option.

A third meta-analysis consisted of 10 RCTs and 19 observational studies (N = 15,862) with patients ages 16 years and older treated operatively vs nonoperatively. Function and return-to-activity rates in both the short term (≤ 1 year) and long term (> 1 year) were evaluated using the Achilles tendon Total Rupture Score (ATRS).³ Surgical management was associated with decreased re-rupture rates but increased complication rates. However, when the analysis was limited to studies using accelerated functional rehabilitation programs, there was no significant difference in re-rupture rate (RR = 0.26 to 1.37; P = .23). Only 1 observational study found a statistically significant difference in short-term functional outcomes favoring operative management, and no studies found a significant difference in long-term functional outcomes. These functional outcomes were not pooled for statistical analysis due to high interrater variability of the ATRS.

An RCT showed equal “customer satisfaction”

One RCT randomized 61 patients to either surgical or nonsurgical management and followed them for a mean of 15.7 years.⁴ Patient-reported outcomes of function, symptoms, and impact on daily life were measured using various surveys. There was no statistically significant difference in the function and impact on daily life after treatment according to the Short Musculoskeletal Function Assessment or the ATRS (P = .289 and .313, respectively). When assessed using the Net Promoter Score (a single-question metric used in consumer industry to assess whether an individual would recommend the product to others), there was no statistical significance for the patients to recommend one treatment over another: 79% of operatively managed patients vs 87% of nonoperatively managed patients would recommend their treatment to others (P = .225).

Recommendations from others

The American College of Foot and Ankle Surgeons consensus statement finds no difference between operative and nonoperative management with regard to complications, functional outcome, and return to activity long term, when looking at available Level 1 evidence.⁵ They do acknowledge that although some Level III studies suggest operative intervention will return high-­functioning patients to full activity sooner, there should be discussion regarding the risks and complications of both operative and nonoperative management. Patients with increased risk factors for postoperative complications (diabetes, obesity, cigarette smoking) should have special considerations regarding the decision to operate.

Editor’s takeaway

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option. However, we cannot say if it is better or worse than operative treatment, because both options have advantages and disadvantages. One must weigh the alternatives with individual patient preferences and circumstances.

Evidence summary

Surgical repair: Re-injury risk goes down, complications risk goes up

A 2021 network meta-analysis including 38 RCTs (N = 2480) reported outcomes in patients ages 18 and older with acute Achilles tendon rupture (AATR) and 3 or more months of follow-up.¹ A significant increase in re-rupture rate was shown in patients who underwent nonoperative vs open repair (risk ratio [RR] = 2.41; 95% CI, 1.12-5.18). There was a significant decrease in wound-related complications in nonoperative vs open-repair patients (RR = 0.23; 95% CI, 0.06-0.88). There was also a significant difference in incidence of sural nerve injury in nonoperative vs open repair (RR = 0.27; 95% CI, 0.08-0.94). There were no significant differences in return to sport between open repair and nonoperative repair (RR = 0.62; 95% CI, 0.22-1.77). Insufficient data were reported to calculate the number needed to treat (NNT) and number needed to harm (NNH) for these outcomes.

Additionally, the authors looked at traditional standard rehabilitation and accelerated functional rehabilitation in both the operative and the nonoperative setting. The type of rehabilitation program did not have a significant impact on complications of re-rupture, wound, or sural nerve injury.

The included studies had an overall low risk of publication bias based on Begg’s funnel plot test (Pr > |z| = 0.86). The highest risk was performance bias, as neither the participants nor personnel were blinded to treatment in 71% of the studies.

Functional outcomes are similar for surgical vs nonoperative repair

In a 2019 meta-analysis of 9 RCTs (N = 822), adults ages 18 and older with AATR and a minimum of 12 months’ follow-up were randomized to either operative or nonoperative repair. There was a decreased rate of rupture with surgical repair and an associated increased rate of complications (ie, superficial wound infections and nerve injury). However, there was no significant difference in Physical Activity Scale (PAS) score between the 2 groups (mean difference, –0.05; 95% CI, –0.37 to 0.27).² With surgical intervention, the NNT for Achilles tendon re-rupture was 15, and the NNH for superficial wound infection and nerve injury, respectively, were 22 and 28. Limitations of the study included different operative techniques and rehab protocols, which may have affected the results of the included studies.

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option.

A third meta-analysis consisted of 10 RCTs and 19 observational studies (N = 15,862) with patients ages 16 years and older treated operatively vs nonoperatively. Function and return-to-activity rates in both the short term (≤ 1 year) and long term (> 1 year) were evaluated using the Achilles tendon Total Rupture Score (ATRS).³ Surgical management was associated with decreased re-rupture rates but increased complication rates. However, when the analysis was limited to studies using accelerated functional rehabilitation programs, there was no significant difference in re-rupture rate (RR = 0.26 to 1.37; P = .23). Only 1 observational study found a statistically significant difference in short-term functional outcomes favoring operative management, and no studies found a significant difference in long-term functional outcomes. These functional outcomes were not pooled for statistical analysis due to high interrater variability of the ATRS.

An RCT showed equal “customer satisfaction”

One RCT randomized 61 patients to either surgical or nonsurgical management and followed them for a mean of 15.7 years.⁴ Patient-reported outcomes of function, symptoms, and impact on daily life were measured using various surveys. There was no statistically significant difference in the function and impact on daily life after treatment according to the Short Musculoskeletal Function Assessment or the ATRS (P = .289 and .313, respectively). When assessed using the Net Promoter Score (a single-question metric used in consumer industry to assess whether an individual would recommend the product to others), there was no statistical significance for the patients to recommend one treatment over another: 79% of operatively managed patients vs 87% of nonoperatively managed patients would recommend their treatment to others (P = .225).

Recommendations from others

The American College of Foot and Ankle Surgeons consensus statement finds no difference between operative and nonoperative management with regard to complications, functional outcome, and return to activity long term, when looking at available Level 1 evidence.⁵ They do acknowledge that although some Level III studies suggest operative intervention will return high-­functioning patients to full activity sooner, there should be discussion regarding the risks and complications of both operative and nonoperative management. Patients with increased risk factors for postoperative complications (diabetes, obesity, cigarette smoking) should have special considerations regarding the decision to operate.

Editor’s takeaway

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option. However, we cannot say if it is better or worse than operative treatment, because both options have advantages and disadvantages. One must weigh the alternatives with individual patient preferences and circumstances.

Evidence summary

Surgical repair: Re-injury risk goes down, complications risk goes up

A 2021 network meta-analysis including 38 RCTs (N = 2480) reported outcomes in patients ages 18 and older with acute Achilles tendon rupture (AATR) and 3 or more months of follow-up.¹ A significant increase in re-rupture rate was shown in patients who underwent nonoperative vs open repair (risk ratio [RR] = 2.41; 95% CI, 1.12-5.18). There was a significant decrease in wound-related complications in nonoperative vs open-repair patients (RR = 0.23; 95% CI, 0.06-0.88). There was also a significant difference in incidence of sural nerve injury in nonoperative vs open repair (RR = 0.27; 95% CI, 0.08-0.94). There were no significant differences in return to sport between open repair and nonoperative repair (RR = 0.62; 95% CI, 0.22-1.77). Insufficient data were reported to calculate the number needed to treat (NNT) and number needed to harm (NNH) for these outcomes.

Additionally, the authors looked at traditional standard rehabilitation and accelerated functional rehabilitation in both the operative and the nonoperative setting. The type of rehabilitation program did not have a significant impact on complications of re-rupture, wound, or sural nerve injury.

The included studies had an overall low risk of publication bias based on Begg’s funnel plot test (Pr > |z| = 0.86). The highest risk was performance bias, as neither the participants nor personnel were blinded to treatment in 71% of the studies.

Functional outcomes are similar for surgical vs nonoperative repair

In a 2019 meta-analysis of 9 RCTs (N = 822), adults ages 18 and older with AATR and a minimum of 12 months’ follow-up were randomized to either operative or nonoperative repair. There was a decreased rate of rupture with surgical repair and an associated increased rate of complications (ie, superficial wound infections and nerve injury). However, there was no significant difference in Physical Activity Scale (PAS) score between the 2 groups (mean difference, –0.05; 95% CI, –0.37 to 0.27).² With surgical intervention, the NNT for Achilles tendon re-rupture was 15, and the NNH for superficial wound infection and nerve injury, respectively, were 22 and 28. Limitations of the study included different operative techniques and rehab protocols, which may have affected the results of the included studies.

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option.

A third meta-analysis consisted of 10 RCTs and 19 observational studies (N = 15,862) with patients ages 16 years and older treated operatively vs nonoperatively. Function and return-to-activity rates in both the short term (≤ 1 year) and long term (> 1 year) were evaluated using the Achilles tendon Total Rupture Score (ATRS).³ Surgical management was associated with decreased re-rupture rates but increased complication rates. However, when the analysis was limited to studies using accelerated functional rehabilitation programs, there was no significant difference in re-rupture rate (RR = 0.26 to 1.37; P = .23). Only 1 observational study found a statistically significant difference in short-term functional outcomes favoring operative management, and no studies found a significant difference in long-term functional outcomes. These functional outcomes were not pooled for statistical analysis due to high interrater variability of the ATRS.

An RCT showed equal “customer satisfaction”

One RCT randomized 61 patients to either surgical or nonsurgical management and followed them for a mean of 15.7 years.⁴ Patient-reported outcomes of function, symptoms, and impact on daily life were measured using various surveys. There was no statistically significant difference in the function and impact on daily life after treatment according to the Short Musculoskeletal Function Assessment or the ATRS (P = .289 and .313, respectively). When assessed using the Net Promoter Score (a single-question metric used in consumer industry to assess whether an individual would recommend the product to others), there was no statistical significance for the patients to recommend one treatment over another: 79% of operatively managed patients vs 87% of nonoperatively managed patients would recommend their treatment to others (P = .225).

Recommendations from others

The American College of Foot and Ankle Surgeons consensus statement finds no difference between operative and nonoperative management with regard to complications, functional outcome, and return to activity long term, when looking at available Level 1 evidence.⁵ They do acknowledge that although some Level III studies suggest operative intervention will return high-­functioning patients to full activity sooner, there should be discussion regarding the risks and complications of both operative and nonoperative management. Patients with increased risk factors for postoperative complications (diabetes, obesity, cigarette smoking) should have special considerations regarding the decision to operate.

Editor’s takeaway

Large data sets with consistent results show that nonoperative treatment of Achilles tendon rupture is an excellent option. However, we cannot say if it is better or worse than operative treatment, because both options have advantages and disadvantages. One must weigh the alternatives with individual patient preferences and circumstances.

Evidence summary

Symptoms improve with PRP—but not significantly

A 2014 double-blind RCT (n = 23) explored recovery outcomes in patients with patellar tendinopathy who received either 1 injection of leukocyte-rich PRP or ultrasound-­guided dry needling.¹ Both groups also completed standardized eccentric exercises. Participants were predominantly men, ages ≥ 18 years. Symptomatic improvement was assessed using the Victorian Institute of Sport ­Assessment–Patella (VISA-P), an 8-item subjective questionnaire of functionality with a range of 0 to 100, with 100 as the maximum score for an asymptomatic individual.

At 12 weeks posttreatment, VISA-P scores improved in both groups. However, the improvement in the dry needling group was not statistically significant (5.2 points; 95% CI, –2.2 to 12.6; P = .20), while in the PRP group it was statistically significant (25.4 points; 95% CI, 10.3 to 40.6; P = .01). At ≥ 26 weeks, statistically significant improvement was observed in both treatment groups: scores improved by 33.2 points (95% CI, 24.1 to 42.4; P = .001) in the dry needling group and by 28.9 points (95% CI, 11.4 to 46.3; P = .01) in the PRP group. However, the difference between the groups’ VISA-P scores at ≥ 26 weeks was not significant (P = .66).¹

No significant differences observed for PRP vs placebo or physical therapy

A 2019 single-blind RCT (n = 57) involved patients who were treated with 1 injection of either leukocyte-rich PRP, leukocyte-poor PRP, or saline, all in combination with 6 weeks of physical therapy.² Participants were predominantly men, ages 18 to 50 years, and engaged in recreational sporting activities. There was no statistically significant difference in mean change in VISA-P score at any timepoint of the 2-year study period. P values were not reported.²

A 2010 RCT (n = 31) compared PRP (unspecified whether leukocyte-rich or -poor) in combination with physical therapy to physical therapy alone.³ Groups were matched for sex, age, and sports activity level; patients in the PRP group were required to have failed previous treatment, while control subjects must not have received any treatment for at least 2 months. Subjects were evaluated pretreatment, immediately posttreatment, and 6 months posttreatment. Clinical evaluation was aided by use of the Tegner activity score, a 1-item score that grades activity level on a scale of 0 to 10; the EuroQol-visual analog scale (EQ-VAS), which evaluates subjective rating of overall health; and pain level scores.

At 6 months posttreatment, no statistically significant differences were observed between groups in EQ-VAS and pain level scores. However, Tegner activity scores among PRP recipients showed significant percent improvement over controls at 6 months posttreatment (39% vs 20%; P = .048).³

Recommendations from others

Currently, national orthopedic and professional athletic medical associations have recommended that further research be conducted in order to make a strong statement in favor of or against PRP.^4,5

Editor’s takeaway

Existing data regarding PRP fails, again, to show consistent benefits. These small sample sizes, inconsistent comparators, and heterogeneous results limit our certainty. This lack of quality evidence does not prove a lack of effect, but it raises serious doubts.

Evidence summary

Symptoms improve with PRP—but not significantly

A 2014 double-blind RCT (n = 23) explored recovery outcomes in patients with patellar tendinopathy who received either 1 injection of leukocyte-rich PRP or ultrasound-­guided dry needling.¹ Both groups also completed standardized eccentric exercises. Participants were predominantly men, ages ≥ 18 years. Symptomatic improvement was assessed using the Victorian Institute of Sport ­Assessment–Patella (VISA-P), an 8-item subjective questionnaire of functionality with a range of 0 to 100, with 100 as the maximum score for an asymptomatic individual.

At 12 weeks posttreatment, VISA-P scores improved in both groups. However, the improvement in the dry needling group was not statistically significant (5.2 points; 95% CI, –2.2 to 12.6; P = .20), while in the PRP group it was statistically significant (25.4 points; 95% CI, 10.3 to 40.6; P = .01). At ≥ 26 weeks, statistically significant improvement was observed in both treatment groups: scores improved by 33.2 points (95% CI, 24.1 to 42.4; P = .001) in the dry needling group and by 28.9 points (95% CI, 11.4 to 46.3; P = .01) in the PRP group. However, the difference between the groups’ VISA-P scores at ≥ 26 weeks was not significant (P = .66).¹

No significant differences observed for PRP vs placebo or physical therapy

A 2019 single-blind RCT (n = 57) involved patients who were treated with 1 injection of either leukocyte-rich PRP, leukocyte-poor PRP, or saline, all in combination with 6 weeks of physical therapy.² Participants were predominantly men, ages 18 to 50 years, and engaged in recreational sporting activities. There was no statistically significant difference in mean change in VISA-P score at any timepoint of the 2-year study period. P values were not reported.²

A 2010 RCT (n = 31) compared PRP (unspecified whether leukocyte-rich or -poor) in combination with physical therapy to physical therapy alone.³ Groups were matched for sex, age, and sports activity level; patients in the PRP group were required to have failed previous treatment, while control subjects must not have received any treatment for at least 2 months. Subjects were evaluated pretreatment, immediately posttreatment, and 6 months posttreatment. Clinical evaluation was aided by use of the Tegner activity score, a 1-item score that grades activity level on a scale of 0 to 10; the EuroQol-visual analog scale (EQ-VAS), which evaluates subjective rating of overall health; and pain level scores.

At 6 months posttreatment, no statistically significant differences were observed between groups in EQ-VAS and pain level scores. However, Tegner activity scores among PRP recipients showed significant percent improvement over controls at 6 months posttreatment (39% vs 20%; P = .048).³

Recommendations from others

Currently, national orthopedic and professional athletic medical associations have recommended that further research be conducted in order to make a strong statement in favor of or against PRP.^4,5

Editor’s takeaway

Existing data regarding PRP fails, again, to show consistent benefits. These small sample sizes, inconsistent comparators, and heterogeneous results limit our certainty. This lack of quality evidence does not prove a lack of effect, but it raises serious doubts.

Evidence summary

Symptoms improve with PRP—but not significantly

A 2014 double-blind RCT (n = 23) explored recovery outcomes in patients with patellar tendinopathy who received either 1 injection of leukocyte-rich PRP or ultrasound-­guided dry needling.¹ Both groups also completed standardized eccentric exercises. Participants were predominantly men, ages ≥ 18 years. Symptomatic improvement was assessed using the Victorian Institute of Sport ­Assessment–Patella (VISA-P), an 8-item subjective questionnaire of functionality with a range of 0 to 100, with 100 as the maximum score for an asymptomatic individual.

At 12 weeks posttreatment, VISA-P scores improved in both groups. However, the improvement in the dry needling group was not statistically significant (5.2 points; 95% CI, –2.2 to 12.6; P = .20), while in the PRP group it was statistically significant (25.4 points; 95% CI, 10.3 to 40.6; P = .01). At ≥ 26 weeks, statistically significant improvement was observed in both treatment groups: scores improved by 33.2 points (95% CI, 24.1 to 42.4; P = .001) in the dry needling group and by 28.9 points (95% CI, 11.4 to 46.3; P = .01) in the PRP group. However, the difference between the groups’ VISA-P scores at ≥ 26 weeks was not significant (P = .66).¹

No significant differences observed for PRP vs placebo or physical therapy

A 2019 single-blind RCT (n = 57) involved patients who were treated with 1 injection of either leukocyte-rich PRP, leukocyte-poor PRP, or saline, all in combination with 6 weeks of physical therapy.² Participants were predominantly men, ages 18 to 50 years, and engaged in recreational sporting activities. There was no statistically significant difference in mean change in VISA-P score at any timepoint of the 2-year study period. P values were not reported.²

A 2010 RCT (n = 31) compared PRP (unspecified whether leukocyte-rich or -poor) in combination with physical therapy to physical therapy alone.³ Groups were matched for sex, age, and sports activity level; patients in the PRP group were required to have failed previous treatment, while control subjects must not have received any treatment for at least 2 months. Subjects were evaluated pretreatment, immediately posttreatment, and 6 months posttreatment. Clinical evaluation was aided by use of the Tegner activity score, a 1-item score that grades activity level on a scale of 0 to 10; the EuroQol-visual analog scale (EQ-VAS), which evaluates subjective rating of overall health; and pain level scores.

At 6 months posttreatment, no statistically significant differences were observed between groups in EQ-VAS and pain level scores. However, Tegner activity scores among PRP recipients showed significant percent improvement over controls at 6 months posttreatment (39% vs 20%; P = .048).³

Recommendations from others

Currently, national orthopedic and professional athletic medical associations have recommended that further research be conducted in order to make a strong statement in favor of or against PRP.^4,5

Editor’s takeaway

Existing data regarding PRP fails, again, to show consistent benefits. These small sample sizes, inconsistent comparators, and heterogeneous results limit our certainty. This lack of quality evidence does not prove a lack of effect, but it raises serious doubts.

Evidence summary

Bicarbonate therapy demonstrates benefit in 2 meta-analyses

Two recent meta-analyses evaluated studies of bicarbonate therapy in patients with CKD, and both found benefit.^1,2

A 2020 meta-analysis included 15 RCTs (N = 2445) of adults (mean age, 61 years; range, 40.5-73.9 years) with CKD.¹ Most trials enrolled patients with an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m²; however, 1 study (N = 80) enrolled patients who had an eGFR of 60 to 90 mL/min/1.73 m² and albuminuria, and another (N = 74) enrolled patients with an eGFR of 15 to 89 mL/min/1.73 m². Four studies included patients with normal baseline bicarbonate levels, while the rest enrolled patients with metabolic acidosis. The primary outcome was CKD progression at study conclusion, which ranged from 3 to 60 months (median, 12 months).

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function (defined by eGFR or creatinine clearance) from baseline to trial completion (14 trials, N = 2073; standardized mean difference [SMD] = 0.26; 95% CI, 0.13-0.40; P = .018; I² = 50%).¹Sodium bicarbonate therapy also resulted in a moderate reduction in the risk of end-stage renal disease (7 trials, N = 1526; risk ratio [RR] = 0.53; 95% CI, 0.30-0.89; P = .011; I² = 69%; number needed to treat [NNT] = 14).¹ There was no difference in hospitalizations for heart failure, risk of worsening blood pressure, or all-cause mortality between the sodium bicarbonate and control groups.

Subgroup analysis by follow-up time found a significant preservation of eGFR only in studies with follow-up > 12 months (4 trials, N = 392; weighted mean difference = 3.71 mL/min/1.73 m²; 95% CI, 0.18-7.24; P = .042; I² = 63%).¹ Duration of therapy did not affect initiation of dialysis. Another subgroup analysis found that low- and moderate-quality studies were more likely than high-quality studies to find a change in the primary outcome. Overall, there was significant heterogeneity among the trials (control intervention, follow-up duration, methods of assessment of kidney function, dosage of sodium bicarbonate), as well as underrepresentation of female, pediatric, and elderly patients.

Another meta-analysis, published in 2019 by a different research group, analyzed 7 RCTs (N = 815) that comprised a subset of those in the newer analysis.² The 2019 analysis similarly found that, compared to placebo or usual care, oral bicarbonate therapy resulted in statistically significantly higher eGFRs at 3 to 60 months’ follow-up (mean difference = 3.1 mL/min/1.73 m²; 95% CI, 1.3-4.9).² The authors noted that the protective effect on eGFR was not seen in studies reporting outcomes at 1 year. Progression to end-stage renal disease or initiation of dialysis were not used as outcomes.

Significant outcomes seen in 1 large study

The largest study (N = 740) included in the 2020 meta-analysis (and discussed separately due to its size and duration) was a multicenter, unblinded, pragmatic trial investigating bicarbonate therapy in CKD.³ Patients were adults (mean age, 67.8 years) with CKD stages 3 to 5 and metabolic acidosis (serum bicarbonate level of 18-24 mmol/L); mean serum creatinine was 2.3 mg/dL, and mean serum bicarbonate was 21.5 mmol/L. Patients with severe heart failure or uncontrolled hypertension were excluded.

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function.

Researchers randomized patients to oral sodium bicarbonate (titrated to a target serum concentration of 24-28 mmol/L) or standard care for a median duration of 30 months. The primary endpoint was time to doubling of serum creatinine, and secondary endpoints included all-cause mortality, time to initiation of dialysis, hospitalization rate, and hospital length of stay.

Continue to: Patients treated with...

Patients treated with bicarbonate therapy had a 64% lower risk of doubling their serum creatinine compared to those treated with standard care (hazard ratio [HR] = 0.36; 95% CI, 0.22-0.58; P < .001; NNT = 9.6).³ Bicarbonate therapy also significantly reduced the risk of dialysis (HR = 0.5; 95% CI, 0.31-0.81; P = .005; NNT = 19); all-cause mortality (HR = 0.43; 95% CI, 0.22-0.87; P = .01; NNT = 27); hospitalization rates (34.6% vs 14.2% by end of study in standard care and bicarbonate groups, respectively; P < .001); and hospital length of stay (1160 total d/y vs 400 total d/y; P < .0001).³ Inspection of Kaplan Meier curves shows outcomes beginning to diverge after 1 to 2 years of treatment. This trial was limited by the lack of blinding, placebo control, and standardization of care protocols.

Recommendations from others

The National Kidney Foundation’s 2012 Kidney Disease Outcomes Quality Initiative guidelines for the management of CKD recommend oral bicarbonate therapy for patients with CKD and serum bicarbonate concentrations < 22 mmol/L.⁴ The guidelines state that serum bicarbonate levels < 22 mmol/L correlate with an increased risk of CKD progression and death, whereas high bicarbonate levels (> 32 mmol/L) correlate with increased risk of death independent of level of kidney function. These guidelines cite small studies of alkali therapy slowing progression of CKD, although it was noted that the evidence base was not strong.

Editor’s takeaway

The evidence shows a small but consistent effect of bicarbonate therapy on CKD progression. For patients with CKD stages 3 to 5 and metabolic acidosis (defined by serum bicarbonate levels < 22 mmol/L), the use of supplemental oral sodium bicarbonate, which is inexpensive and safe, can delay or prevent progression of serious disease.

Evidence summary

Bicarbonate therapy demonstrates benefit in 2 meta-analyses

Two recent meta-analyses evaluated studies of bicarbonate therapy in patients with CKD, and both found benefit.^1,2

A 2020 meta-analysis included 15 RCTs (N = 2445) of adults (mean age, 61 years; range, 40.5-73.9 years) with CKD.¹ Most trials enrolled patients with an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m²; however, 1 study (N = 80) enrolled patients who had an eGFR of 60 to 90 mL/min/1.73 m² and albuminuria, and another (N = 74) enrolled patients with an eGFR of 15 to 89 mL/min/1.73 m². Four studies included patients with normal baseline bicarbonate levels, while the rest enrolled patients with metabolic acidosis. The primary outcome was CKD progression at study conclusion, which ranged from 3 to 60 months (median, 12 months).

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function (defined by eGFR or creatinine clearance) from baseline to trial completion (14 trials, N = 2073; standardized mean difference [SMD] = 0.26; 95% CI, 0.13-0.40; P = .018; I² = 50%).¹Sodium bicarbonate therapy also resulted in a moderate reduction in the risk of end-stage renal disease (7 trials, N = 1526; risk ratio [RR] = 0.53; 95% CI, 0.30-0.89; P = .011; I² = 69%; number needed to treat [NNT] = 14).¹ There was no difference in hospitalizations for heart failure, risk of worsening blood pressure, or all-cause mortality between the sodium bicarbonate and control groups.

Subgroup analysis by follow-up time found a significant preservation of eGFR only in studies with follow-up > 12 months (4 trials, N = 392; weighted mean difference = 3.71 mL/min/1.73 m²; 95% CI, 0.18-7.24; P = .042; I² = 63%).¹ Duration of therapy did not affect initiation of dialysis. Another subgroup analysis found that low- and moderate-quality studies were more likely than high-quality studies to find a change in the primary outcome. Overall, there was significant heterogeneity among the trials (control intervention, follow-up duration, methods of assessment of kidney function, dosage of sodium bicarbonate), as well as underrepresentation of female, pediatric, and elderly patients.

Another meta-analysis, published in 2019 by a different research group, analyzed 7 RCTs (N = 815) that comprised a subset of those in the newer analysis.² The 2019 analysis similarly found that, compared to placebo or usual care, oral bicarbonate therapy resulted in statistically significantly higher eGFRs at 3 to 60 months’ follow-up (mean difference = 3.1 mL/min/1.73 m²; 95% CI, 1.3-4.9).² The authors noted that the protective effect on eGFR was not seen in studies reporting outcomes at 1 year. Progression to end-stage renal disease or initiation of dialysis were not used as outcomes.

Significant outcomes seen in 1 large study

The largest study (N = 740) included in the 2020 meta-analysis (and discussed separately due to its size and duration) was a multicenter, unblinded, pragmatic trial investigating bicarbonate therapy in CKD.³ Patients were adults (mean age, 67.8 years) with CKD stages 3 to 5 and metabolic acidosis (serum bicarbonate level of 18-24 mmol/L); mean serum creatinine was 2.3 mg/dL, and mean serum bicarbonate was 21.5 mmol/L. Patients with severe heart failure or uncontrolled hypertension were excluded.

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function.

Researchers randomized patients to oral sodium bicarbonate (titrated to a target serum concentration of 24-28 mmol/L) or standard care for a median duration of 30 months. The primary endpoint was time to doubling of serum creatinine, and secondary endpoints included all-cause mortality, time to initiation of dialysis, hospitalization rate, and hospital length of stay.

Continue to: Patients treated with...

Patients treated with bicarbonate therapy had a 64% lower risk of doubling their serum creatinine compared to those treated with standard care (hazard ratio [HR] = 0.36; 95% CI, 0.22-0.58; P < .001; NNT = 9.6).³ Bicarbonate therapy also significantly reduced the risk of dialysis (HR = 0.5; 95% CI, 0.31-0.81; P = .005; NNT = 19); all-cause mortality (HR = 0.43; 95% CI, 0.22-0.87; P = .01; NNT = 27); hospitalization rates (34.6% vs 14.2% by end of study in standard care and bicarbonate groups, respectively; P < .001); and hospital length of stay (1160 total d/y vs 400 total d/y; P < .0001).³ Inspection of Kaplan Meier curves shows outcomes beginning to diverge after 1 to 2 years of treatment. This trial was limited by the lack of blinding, placebo control, and standardization of care protocols.

Recommendations from others

The National Kidney Foundation’s 2012 Kidney Disease Outcomes Quality Initiative guidelines for the management of CKD recommend oral bicarbonate therapy for patients with CKD and serum bicarbonate concentrations < 22 mmol/L.⁴ The guidelines state that serum bicarbonate levels < 22 mmol/L correlate with an increased risk of CKD progression and death, whereas high bicarbonate levels (> 32 mmol/L) correlate with increased risk of death independent of level of kidney function. These guidelines cite small studies of alkali therapy slowing progression of CKD, although it was noted that the evidence base was not strong.

Editor’s takeaway

The evidence shows a small but consistent effect of bicarbonate therapy on CKD progression. For patients with CKD stages 3 to 5 and metabolic acidosis (defined by serum bicarbonate levels < 22 mmol/L), the use of supplemental oral sodium bicarbonate, which is inexpensive and safe, can delay or prevent progression of serious disease.

Evidence summary

Bicarbonate therapy demonstrates benefit in 2 meta-analyses

Two recent meta-analyses evaluated studies of bicarbonate therapy in patients with CKD, and both found benefit.^1,2

A 2020 meta-analysis included 15 RCTs (N = 2445) of adults (mean age, 61 years; range, 40.5-73.9 years) with CKD.¹ Most trials enrolled patients with an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m²; however, 1 study (N = 80) enrolled patients who had an eGFR of 60 to 90 mL/min/1.73 m² and albuminuria, and another (N = 74) enrolled patients with an eGFR of 15 to 89 mL/min/1.73 m². Four studies included patients with normal baseline bicarbonate levels, while the rest enrolled patients with metabolic acidosis. The primary outcome was CKD progression at study conclusion, which ranged from 3 to 60 months (median, 12 months).

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function (defined by eGFR or creatinine clearance) from baseline to trial completion (14 trials, N = 2073; standardized mean difference [SMD] = 0.26; 95% CI, 0.13-0.40; P = .018; I² = 50%).¹Sodium bicarbonate therapy also resulted in a moderate reduction in the risk of end-stage renal disease (7 trials, N = 1526; risk ratio [RR] = 0.53; 95% CI, 0.30-0.89; P = .011; I² = 69%; number needed to treat [NNT] = 14).¹ There was no difference in hospitalizations for heart failure, risk of worsening blood pressure, or all-cause mortality between the sodium bicarbonate and control groups.

Subgroup analysis by follow-up time found a significant preservation of eGFR only in studies with follow-up > 12 months (4 trials, N = 392; weighted mean difference = 3.71 mL/min/1.73 m²; 95% CI, 0.18-7.24; P = .042; I² = 63%).¹ Duration of therapy did not affect initiation of dialysis. Another subgroup analysis found that low- and moderate-quality studies were more likely than high-quality studies to find a change in the primary outcome. Overall, there was significant heterogeneity among the trials (control intervention, follow-up duration, methods of assessment of kidney function, dosage of sodium bicarbonate), as well as underrepresentation of female, pediatric, and elderly patients.

Another meta-analysis, published in 2019 by a different research group, analyzed 7 RCTs (N = 815) that comprised a subset of those in the newer analysis.² The 2019 analysis similarly found that, compared to placebo or usual care, oral bicarbonate therapy resulted in statistically significantly higher eGFRs at 3 to 60 months’ follow-up (mean difference = 3.1 mL/min/1.73 m²; 95% CI, 1.3-4.9).² The authors noted that the protective effect on eGFR was not seen in studies reporting outcomes at 1 year. Progression to end-stage renal disease or initiation of dialysis were not used as outcomes.

Significant outcomes seen in 1 large study

The largest study (N = 740) included in the 2020 meta-analysis (and discussed separately due to its size and duration) was a multicenter, unblinded, pragmatic trial investigating bicarbonate therapy in CKD.³ Patients were adults (mean age, 67.8 years) with CKD stages 3 to 5 and metabolic acidosis (serum bicarbonate level of 18-24 mmol/L); mean serum creatinine was 2.3 mg/dL, and mean serum bicarbonate was 21.5 mmol/L. Patients with severe heart failure or uncontrolled hypertension were excluded.

Compared to placebo or no therapy, sodium bicarbonate (variously dosed) resulted in a small reduction in the rate of loss of kidney function.

Researchers randomized patients to oral sodium bicarbonate (titrated to a target serum concentration of 24-28 mmol/L) or standard care for a median duration of 30 months. The primary endpoint was time to doubling of serum creatinine, and secondary endpoints included all-cause mortality, time to initiation of dialysis, hospitalization rate, and hospital length of stay.

Continue to: Patients treated with...

Patients treated with bicarbonate therapy had a 64% lower risk of doubling their serum creatinine compared to those treated with standard care (hazard ratio [HR] = 0.36; 95% CI, 0.22-0.58; P < .001; NNT = 9.6).³ Bicarbonate therapy also significantly reduced the risk of dialysis (HR = 0.5; 95% CI, 0.31-0.81; P = .005; NNT = 19); all-cause mortality (HR = 0.43; 95% CI, 0.22-0.87; P = .01; NNT = 27); hospitalization rates (34.6% vs 14.2% by end of study in standard care and bicarbonate groups, respectively; P < .001); and hospital length of stay (1160 total d/y vs 400 total d/y; P < .0001).³ Inspection of Kaplan Meier curves shows outcomes beginning to diverge after 1 to 2 years of treatment. This trial was limited by the lack of blinding, placebo control, and standardization of care protocols.

Recommendations from others

The National Kidney Foundation’s 2012 Kidney Disease Outcomes Quality Initiative guidelines for the management of CKD recommend oral bicarbonate therapy for patients with CKD and serum bicarbonate concentrations < 22 mmol/L.⁴ The guidelines state that serum bicarbonate levels < 22 mmol/L correlate with an increased risk of CKD progression and death, whereas high bicarbonate levels (> 32 mmol/L) correlate with increased risk of death independent of level of kidney function. These guidelines cite small studies of alkali therapy slowing progression of CKD, although it was noted that the evidence base was not strong.

Editor’s takeaway

The evidence shows a small but consistent effect of bicarbonate therapy on CKD progression. For patients with CKD stages 3 to 5 and metabolic acidosis (defined by serum bicarbonate levels < 22 mmol/L), the use of supplemental oral sodium bicarbonate, which is inexpensive and safe, can delay or prevent progression of serious disease.

Evidence summary

Time to delivery is shortened with combined therapy

Two recent high-quality meta-analyses investigated the effect of adding oxytocin to transcervical Foley balloon placement for cervical dilation. A network meta-analysis, including 30 RCTs (with 6465 pregnant patients), examined the efficacy of multiple combinations of cervical ripening methods.¹ A subset of 7 trials (n = 1313) compared oxytocin infusion with transcervical Foley (inflated to 30-60 mL) to Foley alone. Patients were at > 24 weeks’ gestation with a live fetus and undergoing elective or medical induction of labor; exclusion criteria were standard contraindications to vaginal delivery.

Compared to Foley alone, Foley plus oxytocin reduced both the time to the primary outcome of vaginal delivery (mean duration [MD] = –4.2 h; 95% CI, –1.9 to –6.5) and the time to overall (vaginal and cesarean) delivery (MD = –3.1 h; 95% CI, –1.5 to –4.6). There were no differences in rates of cesarean section, chorioamnionitis, epidural use, or neonatal intensive care unit admission. This analysis did not stratify by parity.¹

In a standard meta-analysis, researchers identified 6 RCTs (N = 1133) comparing transcervical Foley balloon and oxytocin to Foley balloon alone for cervical ripening in pregnant patients at > 23 weeks’ gestation (1 trial was limited to patients at > 37 weeks’ gestation).² Foley balloons were inflated with 30 to 60 mL saline, and oxytocin infusions started at 1 to 2 mU/min and were titrated up to 10 to 40 mU/min. Balloon time was usually 12 hours, but not always stated.

The authors found no statistically significant difference in cesarean rates (the primary outcome) between Foley plus oxytocin vs Foley alone (relative risk [RR] = 0.91; 95% CI, 0.76-1.1). Overall delivery within 12 hours was more likely with combined therapy (RR of remaining pregnant = 0.46; 95% CI, 0.34-0.63), but delivery at 24 hours was not (RR = 0.94; 95% CI, 0.92-1.05). However, in a sub-analysis by parity, nulliparous women who received combined therapy had higher overall delivery rates in 24 hours than did multiparous women (RR = 0.77; 95% CI, 0.62-0.97).²

Adding oxytocin may allow shorter transcervical balloon times

One recent RCT (N = 177) compared labor induction with oxytocin and a single trans-cervical balloon (Cook catheter with only the intrauterine balloon inflated) removed at either 6 or 12 hours.³ Patients were pregnant women (mean age, 31 years) with a term singleton vertex pregnancy, a Bishop score ≤ 6, and no contraindications to vaginal delivery. All patients received a balloon inflated to 60 mL with an oxytocin infusion (2-30 mU/min). The intervention group had the balloon removed at 6 hours, while the control group had it removed at 12 hours.

The mean Bishop score changed by 6 points in each group. Time to overall delivery (the primary outcome) was significantly shorter with 6 hours of balloon time than with 12 hours (19.2 vs 24.3 h; P < .04). Overall delivery within 24 hours was also significantly more likely in the 6-hour group (67.4% vs 47.4%; P < .01), although vaginal delivery in 24 hours did not change (74% vs 59%; P = .07). No differences were seen in cesarean delivery rates or maternal or neonatal morbidity rates.

A look at fixed-dose vs titrated oxytocin

Another RCT (N = 116) examined the effectiveness of cervical ripening using a Foley balloon plus either fixed-dose or titrated low-dose oxytocin.⁴ Patients (mean age, 26 years) had singleton pregnancies at ≥ 37 weeks’ gestation with a Bishop score < 6 and presented for induction of labor. Foley balloons were inflated to 30 mL, and patients received either a fixed oxytocin infusion of 2 mU/min or a titrated infusion starting at 1 mU/min, increasing by 2 mU/min every 30 minutes to a maximum of 20 mU/min.

Continue to: Thre was no statistically...

High-quality evidence shows that the addition of oxytocin to balloon cervical ripening shortens the time to delivery.

There was no statistically significant difference in median time from Foley placement to overall delivery (the primary outcome) between the fixed low-dose and incremental low-dose groups in either nulliparous women (24 vs 19 h; P = .18) or multiparous women (16 vs 12 h; P = .68). The authors acknowledged the study may have been underpowered to detect a true difference.

Recommendations from others

A 2009 Practice Bulletin from the American College of Obstetricians and Gynecologists (ACOG) recommended the Foley catheter as a reasonable and effective alternative to prostaglandins for cervical ripening and the induction of labor (based on good-quality evidence).⁵ The guideline stated that Foley catheter placement before oxytocin induction reduced both the duration of labor and risk of cesarean delivery, but that the use of oxytocin along with a Foley catheter did not appear to shorten the time to delivery.

Editor’s takeaway

High-quality evidence shows us that the addition of oxytocin to balloon cervical ripening shortens the time to delivery. This newer evidence may prompt an update to the 2009 ACOG statement.

Evidence summary

Time to delivery is shortened with combined therapy

Two recent high-quality meta-analyses investigated the effect of adding oxytocin to transcervical Foley balloon placement for cervical dilation. A network meta-analysis, including 30 RCTs (with 6465 pregnant patients), examined the efficacy of multiple combinations of cervical ripening methods.¹ A subset of 7 trials (n = 1313) compared oxytocin infusion with transcervical Foley (inflated to 30-60 mL) to Foley alone. Patients were at > 24 weeks’ gestation with a live fetus and undergoing elective or medical induction of labor; exclusion criteria were standard contraindications to vaginal delivery.

Compared to Foley alone, Foley plus oxytocin reduced both the time to the primary outcome of vaginal delivery (mean duration [MD] = –4.2 h; 95% CI, –1.9 to –6.5) and the time to overall (vaginal and cesarean) delivery (MD = –3.1 h; 95% CI, –1.5 to –4.6). There were no differences in rates of cesarean section, chorioamnionitis, epidural use, or neonatal intensive care unit admission. This analysis did not stratify by parity.¹

In a standard meta-analysis, researchers identified 6 RCTs (N = 1133) comparing transcervical Foley balloon and oxytocin to Foley balloon alone for cervical ripening in pregnant patients at > 23 weeks’ gestation (1 trial was limited to patients at > 37 weeks’ gestation).² Foley balloons were inflated with 30 to 60 mL saline, and oxytocin infusions started at 1 to 2 mU/min and were titrated up to 10 to 40 mU/min. Balloon time was usually 12 hours, but not always stated.

The authors found no statistically significant difference in cesarean rates (the primary outcome) between Foley plus oxytocin vs Foley alone (relative risk [RR] = 0.91; 95% CI, 0.76-1.1). Overall delivery within 12 hours was more likely with combined therapy (RR of remaining pregnant = 0.46; 95% CI, 0.34-0.63), but delivery at 24 hours was not (RR = 0.94; 95% CI, 0.92-1.05). However, in a sub-analysis by parity, nulliparous women who received combined therapy had higher overall delivery rates in 24 hours than did multiparous women (RR = 0.77; 95% CI, 0.62-0.97).²

Adding oxytocin may allow shorter transcervical balloon times

One recent RCT (N = 177) compared labor induction with oxytocin and a single trans-cervical balloon (Cook catheter with only the intrauterine balloon inflated) removed at either 6 or 12 hours.³ Patients were pregnant women (mean age, 31 years) with a term singleton vertex pregnancy, a Bishop score ≤ 6, and no contraindications to vaginal delivery. All patients received a balloon inflated to 60 mL with an oxytocin infusion (2-30 mU/min). The intervention group had the balloon removed at 6 hours, while the control group had it removed at 12 hours.

The mean Bishop score changed by 6 points in each group. Time to overall delivery (the primary outcome) was significantly shorter with 6 hours of balloon time than with 12 hours (19.2 vs 24.3 h; P < .04). Overall delivery within 24 hours was also significantly more likely in the 6-hour group (67.4% vs 47.4%; P < .01), although vaginal delivery in 24 hours did not change (74% vs 59%; P = .07). No differences were seen in cesarean delivery rates or maternal or neonatal morbidity rates.

A look at fixed-dose vs titrated oxytocin

Another RCT (N = 116) examined the effectiveness of cervical ripening using a Foley balloon plus either fixed-dose or titrated low-dose oxytocin.⁴ Patients (mean age, 26 years) had singleton pregnancies at ≥ 37 weeks’ gestation with a Bishop score < 6 and presented for induction of labor. Foley balloons were inflated to 30 mL, and patients received either a fixed oxytocin infusion of 2 mU/min or a titrated infusion starting at 1 mU/min, increasing by 2 mU/min every 30 minutes to a maximum of 20 mU/min.

Continue to: Thre was no statistically...

High-quality evidence shows that the addition of oxytocin to balloon cervical ripening shortens the time to delivery.

There was no statistically significant difference in median time from Foley placement to overall delivery (the primary outcome) between the fixed low-dose and incremental low-dose groups in either nulliparous women (24 vs 19 h; P = .18) or multiparous women (16 vs 12 h; P = .68). The authors acknowledged the study may have been underpowered to detect a true difference.

Recommendations from others

A 2009 Practice Bulletin from the American College of Obstetricians and Gynecologists (ACOG) recommended the Foley catheter as a reasonable and effective alternative to prostaglandins for cervical ripening and the induction of labor (based on good-quality evidence).⁵ The guideline stated that Foley catheter placement before oxytocin induction reduced both the duration of labor and risk of cesarean delivery, but that the use of oxytocin along with a Foley catheter did not appear to shorten the time to delivery.

Editor’s takeaway

High-quality evidence shows us that the addition of oxytocin to balloon cervical ripening shortens the time to delivery. This newer evidence may prompt an update to the 2009 ACOG statement.

Evidence summary

Time to delivery is shortened with combined therapy

Two recent high-quality meta-analyses investigated the effect of adding oxytocin to transcervical Foley balloon placement for cervical dilation. A network meta-analysis, including 30 RCTs (with 6465 pregnant patients), examined the efficacy of multiple combinations of cervical ripening methods.¹ A subset of 7 trials (n = 1313) compared oxytocin infusion with transcervical Foley (inflated to 30-60 mL) to Foley alone. Patients were at > 24 weeks’ gestation with a live fetus and undergoing elective or medical induction of labor; exclusion criteria were standard contraindications to vaginal delivery.

Compared to Foley alone, Foley plus oxytocin reduced both the time to the primary outcome of vaginal delivery (mean duration [MD] = –4.2 h; 95% CI, –1.9 to –6.5) and the time to overall (vaginal and cesarean) delivery (MD = –3.1 h; 95% CI, –1.5 to –4.6). There were no differences in rates of cesarean section, chorioamnionitis, epidural use, or neonatal intensive care unit admission. This analysis did not stratify by parity.¹

In a standard meta-analysis, researchers identified 6 RCTs (N = 1133) comparing transcervical Foley balloon and oxytocin to Foley balloon alone for cervical ripening in pregnant patients at > 23 weeks’ gestation (1 trial was limited to patients at > 37 weeks’ gestation).² Foley balloons were inflated with 30 to 60 mL saline, and oxytocin infusions started at 1 to 2 mU/min and were titrated up to 10 to 40 mU/min. Balloon time was usually 12 hours, but not always stated.

The authors found no statistically significant difference in cesarean rates (the primary outcome) between Foley plus oxytocin vs Foley alone (relative risk [RR] = 0.91; 95% CI, 0.76-1.1). Overall delivery within 12 hours was more likely with combined therapy (RR of remaining pregnant = 0.46; 95% CI, 0.34-0.63), but delivery at 24 hours was not (RR = 0.94; 95% CI, 0.92-1.05). However, in a sub-analysis by parity, nulliparous women who received combined therapy had higher overall delivery rates in 24 hours than did multiparous women (RR = 0.77; 95% CI, 0.62-0.97).²

Adding oxytocin may allow shorter transcervical balloon times

One recent RCT (N = 177) compared labor induction with oxytocin and a single trans-cervical balloon (Cook catheter with only the intrauterine balloon inflated) removed at either 6 or 12 hours.³ Patients were pregnant women (mean age, 31 years) with a term singleton vertex pregnancy, a Bishop score ≤ 6, and no contraindications to vaginal delivery. All patients received a balloon inflated to 60 mL with an oxytocin infusion (2-30 mU/min). The intervention group had the balloon removed at 6 hours, while the control group had it removed at 12 hours.

The mean Bishop score changed by 6 points in each group. Time to overall delivery (the primary outcome) was significantly shorter with 6 hours of balloon time than with 12 hours (19.2 vs 24.3 h; P < .04). Overall delivery within 24 hours was also significantly more likely in the 6-hour group (67.4% vs 47.4%; P < .01), although vaginal delivery in 24 hours did not change (74% vs 59%; P = .07). No differences were seen in cesarean delivery rates or maternal or neonatal morbidity rates.

A look at fixed-dose vs titrated oxytocin

Another RCT (N = 116) examined the effectiveness of cervical ripening using a Foley balloon plus either fixed-dose or titrated low-dose oxytocin.⁴ Patients (mean age, 26 years) had singleton pregnancies at ≥ 37 weeks’ gestation with a Bishop score < 6 and presented for induction of labor. Foley balloons were inflated to 30 mL, and patients received either a fixed oxytocin infusion of 2 mU/min or a titrated infusion starting at 1 mU/min, increasing by 2 mU/min every 30 minutes to a maximum of 20 mU/min.

Continue to: Thre was no statistically...

High-quality evidence shows that the addition of oxytocin to balloon cervical ripening shortens the time to delivery.

There was no statistically significant difference in median time from Foley placement to overall delivery (the primary outcome) between the fixed low-dose and incremental low-dose groups in either nulliparous women (24 vs 19 h; P = .18) or multiparous women (16 vs 12 h; P = .68). The authors acknowledged the study may have been underpowered to detect a true difference.

Recommendations from others

A 2009 Practice Bulletin from the American College of Obstetricians and Gynecologists (ACOG) recommended the Foley catheter as a reasonable and effective alternative to prostaglandins for cervical ripening and the induction of labor (based on good-quality evidence).⁵ The guideline stated that Foley catheter placement before oxytocin induction reduced both the duration of labor and risk of cesarean delivery, but that the use of oxytocin along with a Foley catheter did not appear to shorten the time to delivery.

Editor’s takeaway

High-quality evidence shows us that the addition of oxytocin to balloon cervical ripening shortens the time to delivery. This newer evidence may prompt an update to the 2009 ACOG statement.

Evidence summary

Alone or in combination with hCG, clomiphene citrate is effective

A 2018 multicenter prospective RCT (n = 283) compared the serum testosterone response in men (mean age, 41.8 ± 10.4 years) with hypogonadism before and after treatment with either CC, human chorionic gonadotropin (hCG), or a combination of both therapies.¹ All patients wanted to maintain fertility, had normal follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels, had no history of testosterone therapy, had low (< 300 ng/dL) serum testosterone levels on at least 2 samples, and had at least 3 positive symptoms from the quantitative Androgen Deficiency in the Aging Male questionnaire (qADAM; a 10-item, graded-response tool measuring symptom severity from 1 to 5).

Patients were randomized into either the CC group (50 mg oral; n = 95), the hCG group (5000 IU injections twice weekly; n = 94), or the CC + hCG group (n = 94). Testosterone levels were measured at baseline and at 1 and 3 months after therapy initiation; qADAM questionnaire scores were also recorded but ultimately not used due to concerns with baseline heterogeneity among groups.

Average baseline serum testosterone levels for the CC, hCG, and CC + hCG groups were 243 ng/dL, 222 ng/dL, and 226 ng/dL, respectively. By 3 months, these levels had increased to 548 ng/dL (95% CI, 505-591) in the CC group, 467 ng/dL (95% CI, 440-494) in the hCG group, and 531 ng/dL (95% CI, 492-570) in the CC + hCG group. While there was not a significant difference between the CC and CC + hCG groups at 3 months (P = .579), both groups were superior to the hCG-only group (P = .002 for each).

CC and testosterone gel are comparable; testosterone injection is better

In a 2014 retrospective study, researchers reviewed the charts of 1150 men taking any form of testosterone supplementation therapy (TST). They compared treatment efficacy and qADAM satisfaction scores in 93 age-matched men with symptomatic hypogonadism who were treated with either CC (n = 31), testosterone injections (n = 31), or testosterone topical gel (n = 31).² Eugonadal men not taking TST (n = 31) served as controls.

Inclusion criteria were based on treatment regimens of CC and TST. Participants in the treatment groups had a baseline total testosterone level < 300 ng/dL and had reported ≤ 3 positive symptoms on the qADAM questionnaire. Treatment regimens included CC (25 mg orally once daily), testosterone injections (testosterone cypionate 100 to 200 mg intramuscularly once weekly), and testosterone gel (Testim 1% or AndroGel 1.62%, 2 to 4 pumps/d).

The study results demonstrated an increase in median testosterone from baseline levels in all treatment groups when compared to placebo: CC (from 247 to 504 ng/dL), testosterone injections (from 224 to 1104 ng/dL), and testosterone gels (from 230 to 412 ng/dL) (P < .05). Men receiving testosterone injections had the highest increase in serum testosterone levels (956 ng/dL).

While the final mean serum total testosterone was highest in the testosterone injection group (1014 ng/dL; P < .01), the mean levels for those using CC and those using testosterone gels were comparable (525 ng/dL vs 412 ng/dL). Serum estradiol levels were also higher in men receiving testosterone injections, compared to men using CC, those using testosterone gels, and those not receiving TST (6.0 vs 2.0, 2.0, and 2.0 ng/dL, respectively; each P < .01).

Continue to: The qADAM scores...

The qADAM scores for hypogonadal symptoms showed no significant difference in men treated with CC, testosterone injections, and testosterone gels and those not receiving TST (35, 39, 36, and 34, respectively). Men receiving testosterone injections reported greater libido (range, 1-5) than men using CC, those using testosterone gels, and those not on TST (4 vs 3, 3, and 3; P = .047, .04, and < .01, respectively), but it is uncertain if this is clinically meaningful.

Enclomiphene citrate demonstrates improvement in hormone levels

A 2014 Phase II RCT investigated the effects of oral EC—a trans-isomer of CC—compared to topical testosterone 1% gel (T gel) in 124 men with secondary hypogonadism.³ Entry criteria included a baseline morning total testosterone level of < 250 ng/dL on 2 occasions. Participants were divided into 4 groups: 12.5-mg dose of EC, 25-mg dose of EC, T gel, and placebo.

The EC groups and the T gel group had comparable increases in testosterone levels after 3 months of treatment, without statistical significance. The 3-month change in serum testosterone level from baseline was 217 to 471 ng/dL (95% CI, 399-543) in the 12.5-mg dose group; 209 to 405 ng/dL (95% CI, 349-462) in the 25-mg dose group; and 210 to 462 ng/dL (95% CI, 359-565) in the T gel group. The placebo group had a decrease in serum testosterone levels, from 213 to 198 ng/dL (95% CI, 171-226).

Men with low or low-normal serum luteinizing hormone levels may be good candidates for the use of SERMs for management of testosterone deficiency.

Also, the EC groups demonstrated increases in LH and FSH levels from baseline to 3 months, while the T gel group showed a suppression (to low-normal range) in both levels: LH, 1.4 mIU/mL (decrease of 4.4 mIU/mL) and FSH, 2.4 mIU/mL (decrease of 2.4 mIU/mL). Among a subset of men (n = 67) who had at least 2 assessments at the end of 3 months, the researchers also analyzed changes in sperm concentration, using the lower limit of normal (15 million/mL). The number of men with a low sperm concentration increased significantly in the topical T gel group (16% to 53%) compared to the 12.5-mg EC group (decrease from 16% to 12%; P = .0008) and the 25-mg EC group (decrease from 5% to 0%; P = .0007), as well as compared to the placebo group (increase from 8% to 15%; P = .007).

With EC, testosterone remains elevated after treatment cessation

A 2016 2-center parallel, double-blind, ­placebo-controlled RCT evaluated the effect of 2 doses of EC (12.5 mg and 25 mg; n = 85) vs testosterone gel (1.62%; n = 85) vs placebo (n = 86) on serum testosterone, LH, FSH, and sperm counts in 256 overweight and obese men ages 18 to 60 years who had 2 morning testosterone measurements < 300 ng/dL and a low or inappropriately normal LH level for 16 weeks.⁴ All baseline characteristics, including age, BMI, sperm concentration, and serum total testosterone were statistically consistent within groups at both centers. For men receiving EC who did not achieve a testosterone level > 450 ng/dL, there was an up-titration from 12.5 mg to 25 mg at Week 4.

Continue to: All active treatment groups...

All active treatment groups showed increases in testosterone level during treatment (P < .001); however, FSH and LH levels increased in the EC group and decreased in the testosterone gel group (P < .001). Serum testosterone levels improved to 428.8 ng/dL (95% CI, 395-462) and 368.8 ng/dL (95% CI, 307-431), respectively, in the combined EC and testosterone gel groups at 16 weeks. Of note, total testosterone levels after cessation of treatment (off-drug point) rapidly decreased below baseline in the testosterone gel group compared to the pooled EC group, which remained elevated above baseline for at least 7 days.

Composite end-point analysis was performed, with success considered if men achieved both testosterone in normal range (300-1040 ng/dL) and sperm concentrations ≥ 10 × 10⁶. The pooled data studies showed EC was more successful than testosterone gel in achieving both endpoints (63.5% vs 24.7%; P < .001). No difference in the incidence of treatment-related adverse effects between groups was noted.

There were no major adverse effects, even after 3+ years of treatment

A 2019 retrospective cohort study of 400 men treated for symptomatic hypogonadism with CC sought to determine if improvements in testosterone, hypogonadal symptoms, and adverse effects were similar for those treated for ≤ 3 years (n = 280) and those treated for > 3 years (n = 120).⁵ Outcomes included serum testosterone and estradiol levels, symptom improvement (by qADAM questionnaire), and adverse effects.

All participants had a baseline testosterone level < 300 ng/dL, and all participants received CC therapy. Men received 25 mg/d with titration to 50 mg/d when testosterone did not improve to ≥ 300 ng/dL after 4 weeks.

When comparing outcomes across the 2 groups, there were no significant differences. Serum testosterone levels improved to 579 ng/dL (95% CI, 554-605) and 542 ng/dL (95% CI, 504-580) in the ≤ 3 years and > 3 years groups, respectively. Meanwhile, 79% of men in the ≤ 3 years group reported symptom improvement (improvement in libido, erection, or 3 other of the 10 domains of the qADAM questionnaire), while 77% of those in the > 3 years group reported improvement (P = .60).

Continue to: Finally, the percentage of men reporting...

Finally, the percentage of men reporting adverse effects did not significantly differ between groups: 9% in the ≤ 3 years group and 8% in the > 3 years group (P = .85). The most common adverse effects reported in order of frequency were mood changes, blurred vision, breast tenderness, hypertension, hematocrit changes, and flushing. No major adverse events (eg, myocardial infarction, cerebrovascular accident, venous thromboembolism, suicidal behavior) were reported in any patients.

Both the American Urological Association and the Canadian Urological Association support the use of SERMs, especially in hypogonadal men who are interested in fertility preservation.

Of note, although measured estrogen levels at the end of treatment were similar for both groups (54.8 pg/mL in the ≤ 3 years group vs 54.6 pg/mL in the > 3 years group), 37% of patients treated for > 3 years did receive anastrozole treatment for hyperestrogenism compared to 15% in the ≤ 3 years group (P = .05). The authors caution, though, that due to only 20% of the cohort patients having data on pre- and post-treatment estradiol levels, the study was likely underpowered to detect true differences among subgroups.

Recommendations from others

Current American Urological Association and Canadian Urological Association Guidelines note that while greater study on nontraditional testosterone therapies is needed, both organizations support use of SERMs, especially in hypogonadal men who are interested in fertility preservation, as increases in endogenous serum testosterone production do not impact fertility potential, unlike exogenous hormonal replacement.^6,7 Additionally, men with low or low-normal serum LH levels may also be good candidates for the use of SERMs for management of testosterone deficiency.

Editor’s takeaway

Laboratory data (disease oriented) consistently shows that SERMs effectively increase testosterone levels to those comparable with testosterone gels. SERMs resulted in higher semen counts and maintained LH and FSH levels, but there were instances of hyperestrogenism. Data on longer-term benefits and adverse effects of both SERMs and testosterone supplementation are still needed.

Evidence summary

Alone or in combination with hCG, clomiphene citrate is effective

A 2018 multicenter prospective RCT (n = 283) compared the serum testosterone response in men (mean age, 41.8 ± 10.4 years) with hypogonadism before and after treatment with either CC, human chorionic gonadotropin (hCG), or a combination of both therapies.¹ All patients wanted to maintain fertility, had normal follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels, had no history of testosterone therapy, had low (< 300 ng/dL) serum testosterone levels on at least 2 samples, and had at least 3 positive symptoms from the quantitative Androgen Deficiency in the Aging Male questionnaire (qADAM; a 10-item, graded-response tool measuring symptom severity from 1 to 5).

Patients were randomized into either the CC group (50 mg oral; n = 95), the hCG group (5000 IU injections twice weekly; n = 94), or the CC + hCG group (n = 94). Testosterone levels were measured at baseline and at 1 and 3 months after therapy initiation; qADAM questionnaire scores were also recorded but ultimately not used due to concerns with baseline heterogeneity among groups.

Average baseline serum testosterone levels for the CC, hCG, and CC + hCG groups were 243 ng/dL, 222 ng/dL, and 226 ng/dL, respectively. By 3 months, these levels had increased to 548 ng/dL (95% CI, 505-591) in the CC group, 467 ng/dL (95% CI, 440-494) in the hCG group, and 531 ng/dL (95% CI, 492-570) in the CC + hCG group. While there was not a significant difference between the CC and CC + hCG groups at 3 months (P = .579), both groups were superior to the hCG-only group (P = .002 for each).

CC and testosterone gel are comparable; testosterone injection is better

In a 2014 retrospective study, researchers reviewed the charts of 1150 men taking any form of testosterone supplementation therapy (TST). They compared treatment efficacy and qADAM satisfaction scores in 93 age-matched men with symptomatic hypogonadism who were treated with either CC (n = 31), testosterone injections (n = 31), or testosterone topical gel (n = 31).² Eugonadal men not taking TST (n = 31) served as controls.

Inclusion criteria were based on treatment regimens of CC and TST. Participants in the treatment groups had a baseline total testosterone level < 300 ng/dL and had reported ≤ 3 positive symptoms on the qADAM questionnaire. Treatment regimens included CC (25 mg orally once daily), testosterone injections (testosterone cypionate 100 to 200 mg intramuscularly once weekly), and testosterone gel (Testim 1% or AndroGel 1.62%, 2 to 4 pumps/d).

The study results demonstrated an increase in median testosterone from baseline levels in all treatment groups when compared to placebo: CC (from 247 to 504 ng/dL), testosterone injections (from 224 to 1104 ng/dL), and testosterone gels (from 230 to 412 ng/dL) (P < .05). Men receiving testosterone injections had the highest increase in serum testosterone levels (956 ng/dL).

While the final mean serum total testosterone was highest in the testosterone injection group (1014 ng/dL; P < .01), the mean levels for those using CC and those using testosterone gels were comparable (525 ng/dL vs 412 ng/dL). Serum estradiol levels were also higher in men receiving testosterone injections, compared to men using CC, those using testosterone gels, and those not receiving TST (6.0 vs 2.0, 2.0, and 2.0 ng/dL, respectively; each P < .01).

Continue to: The qADAM scores...

The qADAM scores for hypogonadal symptoms showed no significant difference in men treated with CC, testosterone injections, and testosterone gels and those not receiving TST (35, 39, 36, and 34, respectively). Men receiving testosterone injections reported greater libido (range, 1-5) than men using CC, those using testosterone gels, and those not on TST (4 vs 3, 3, and 3; P = .047, .04, and < .01, respectively), but it is uncertain if this is clinically meaningful.

Enclomiphene citrate demonstrates improvement in hormone levels

A 2014 Phase II RCT investigated the effects of oral EC—a trans-isomer of CC—compared to topical testosterone 1% gel (T gel) in 124 men with secondary hypogonadism.³ Entry criteria included a baseline morning total testosterone level of < 250 ng/dL on 2 occasions. Participants were divided into 4 groups: 12.5-mg dose of EC, 25-mg dose of EC, T gel, and placebo.

The EC groups and the T gel group had comparable increases in testosterone levels after 3 months of treatment, without statistical significance. The 3-month change in serum testosterone level from baseline was 217 to 471 ng/dL (95% CI, 399-543) in the 12.5-mg dose group; 209 to 405 ng/dL (95% CI, 349-462) in the 25-mg dose group; and 210 to 462 ng/dL (95% CI, 359-565) in the T gel group. The placebo group had a decrease in serum testosterone levels, from 213 to 198 ng/dL (95% CI, 171-226).

Men with low or low-normal serum luteinizing hormone levels may be good candidates for the use of SERMs for management of testosterone deficiency.

Also, the EC groups demonstrated increases in LH and FSH levels from baseline to 3 months, while the T gel group showed a suppression (to low-normal range) in both levels: LH, 1.4 mIU/mL (decrease of 4.4 mIU/mL) and FSH, 2.4 mIU/mL (decrease of 2.4 mIU/mL). Among a subset of men (n = 67) who had at least 2 assessments at the end of 3 months, the researchers also analyzed changes in sperm concentration, using the lower limit of normal (15 million/mL). The number of men with a low sperm concentration increased significantly in the topical T gel group (16% to 53%) compared to the 12.5-mg EC group (decrease from 16% to 12%; P = .0008) and the 25-mg EC group (decrease from 5% to 0%; P = .0007), as well as compared to the placebo group (increase from 8% to 15%; P = .007).

With EC, testosterone remains elevated after treatment cessation

A 2016 2-center parallel, double-blind, ­placebo-controlled RCT evaluated the effect of 2 doses of EC (12.5 mg and 25 mg; n = 85) vs testosterone gel (1.62%; n = 85) vs placebo (n = 86) on serum testosterone, LH, FSH, and sperm counts in 256 overweight and obese men ages 18 to 60 years who had 2 morning testosterone measurements < 300 ng/dL and a low or inappropriately normal LH level for 16 weeks.⁴ All baseline characteristics, including age, BMI, sperm concentration, and serum total testosterone were statistically consistent within groups at both centers. For men receiving EC who did not achieve a testosterone level > 450 ng/dL, there was an up-titration from 12.5 mg to 25 mg at Week 4.

Continue to: All active treatment groups...

All active treatment groups showed increases in testosterone level during treatment (P < .001); however, FSH and LH levels increased in the EC group and decreased in the testosterone gel group (P < .001). Serum testosterone levels improved to 428.8 ng/dL (95% CI, 395-462) and 368.8 ng/dL (95% CI, 307-431), respectively, in the combined EC and testosterone gel groups at 16 weeks. Of note, total testosterone levels after cessation of treatment (off-drug point) rapidly decreased below baseline in the testosterone gel group compared to the pooled EC group, which remained elevated above baseline for at least 7 days.

Composite end-point analysis was performed, with success considered if men achieved both testosterone in normal range (300-1040 ng/dL) and sperm concentrations ≥ 10 × 10⁶. The pooled data studies showed EC was more successful than testosterone gel in achieving both endpoints (63.5% vs 24.7%; P < .001). No difference in the incidence of treatment-related adverse effects between groups was noted.

There were no major adverse effects, even after 3+ years of treatment

A 2019 retrospective cohort study of 400 men treated for symptomatic hypogonadism with CC sought to determine if improvements in testosterone, hypogonadal symptoms, and adverse effects were similar for those treated for ≤ 3 years (n = 280) and those treated for > 3 years (n = 120).⁵ Outcomes included serum testosterone and estradiol levels, symptom improvement (by qADAM questionnaire), and adverse effects.

All participants had a baseline testosterone level < 300 ng/dL, and all participants received CC therapy. Men received 25 mg/d with titration to 50 mg/d when testosterone did not improve to ≥ 300 ng/dL after 4 weeks.

When comparing outcomes across the 2 groups, there were no significant differences. Serum testosterone levels improved to 579 ng/dL (95% CI, 554-605) and 542 ng/dL (95% CI, 504-580) in the ≤ 3 years and > 3 years groups, respectively. Meanwhile, 79% of men in the ≤ 3 years group reported symptom improvement (improvement in libido, erection, or 3 other of the 10 domains of the qADAM questionnaire), while 77% of those in the > 3 years group reported improvement (P = .60).

Continue to: Finally, the percentage of men reporting...

Finally, the percentage of men reporting adverse effects did not significantly differ between groups: 9% in the ≤ 3 years group and 8% in the > 3 years group (P = .85). The most common adverse effects reported in order of frequency were mood changes, blurred vision, breast tenderness, hypertension, hematocrit changes, and flushing. No major adverse events (eg, myocardial infarction, cerebrovascular accident, venous thromboembolism, suicidal behavior) were reported in any patients.

Both the American Urological Association and the Canadian Urological Association support the use of SERMs, especially in hypogonadal men who are interested in fertility preservation.

Of note, although measured estrogen levels at the end of treatment were similar for both groups (54.8 pg/mL in the ≤ 3 years group vs 54.6 pg/mL in the > 3 years group), 37% of patients treated for > 3 years did receive anastrozole treatment for hyperestrogenism compared to 15% in the ≤ 3 years group (P = .05). The authors caution, though, that due to only 20% of the cohort patients having data on pre- and post-treatment estradiol levels, the study was likely underpowered to detect true differences among subgroups.

Recommendations from others

Current American Urological Association and Canadian Urological Association Guidelines note that while greater study on nontraditional testosterone therapies is needed, both organizations support use of SERMs, especially in hypogonadal men who are interested in fertility preservation, as increases in endogenous serum testosterone production do not impact fertility potential, unlike exogenous hormonal replacement.^6,7 Additionally, men with low or low-normal serum LH levels may also be good candidates for the use of SERMs for management of testosterone deficiency.

Editor’s takeaway

Laboratory data (disease oriented) consistently shows that SERMs effectively increase testosterone levels to those comparable with testosterone gels. SERMs resulted in higher semen counts and maintained LH and FSH levels, but there were instances of hyperestrogenism. Data on longer-term benefits and adverse effects of both SERMs and testosterone supplementation are still needed.

Evidence summary

Alone or in combination with hCG, clomiphene citrate is effective

A 2018 multicenter prospective RCT (n = 283) compared the serum testosterone response in men (mean age, 41.8 ± 10.4 years) with hypogonadism before and after treatment with either CC, human chorionic gonadotropin (hCG), or a combination of both therapies.¹ All patients wanted to maintain fertility, had normal follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels, had no history of testosterone therapy, had low (< 300 ng/dL) serum testosterone levels on at least 2 samples, and had at least 3 positive symptoms from the quantitative Androgen Deficiency in the Aging Male questionnaire (qADAM; a 10-item, graded-response tool measuring symptom severity from 1 to 5).

Patients were randomized into either the CC group (50 mg oral; n = 95), the hCG group (5000 IU injections twice weekly; n = 94), or the CC + hCG group (n = 94). Testosterone levels were measured at baseline and at 1 and 3 months after therapy initiation; qADAM questionnaire scores were also recorded but ultimately not used due to concerns with baseline heterogeneity among groups.

Average baseline serum testosterone levels for the CC, hCG, and CC + hCG groups were 243 ng/dL, 222 ng/dL, and 226 ng/dL, respectively. By 3 months, these levels had increased to 548 ng/dL (95% CI, 505-591) in the CC group, 467 ng/dL (95% CI, 440-494) in the hCG group, and 531 ng/dL (95% CI, 492-570) in the CC + hCG group. While there was not a significant difference between the CC and CC + hCG groups at 3 months (P = .579), both groups were superior to the hCG-only group (P = .002 for each).

CC and testosterone gel are comparable; testosterone injection is better

In a 2014 retrospective study, researchers reviewed the charts of 1150 men taking any form of testosterone supplementation therapy (TST). They compared treatment efficacy and qADAM satisfaction scores in 93 age-matched men with symptomatic hypogonadism who were treated with either CC (n = 31), testosterone injections (n = 31), or testosterone topical gel (n = 31).² Eugonadal men not taking TST (n = 31) served as controls.

Inclusion criteria were based on treatment regimens of CC and TST. Participants in the treatment groups had a baseline total testosterone level < 300 ng/dL and had reported ≤ 3 positive symptoms on the qADAM questionnaire. Treatment regimens included CC (25 mg orally once daily), testosterone injections (testosterone cypionate 100 to 200 mg intramuscularly once weekly), and testosterone gel (Testim 1% or AndroGel 1.62%, 2 to 4 pumps/d).

The study results demonstrated an increase in median testosterone from baseline levels in all treatment groups when compared to placebo: CC (from 247 to 504 ng/dL), testosterone injections (from 224 to 1104 ng/dL), and testosterone gels (from 230 to 412 ng/dL) (P < .05). Men receiving testosterone injections had the highest increase in serum testosterone levels (956 ng/dL).

While the final mean serum total testosterone was highest in the testosterone injection group (1014 ng/dL; P < .01), the mean levels for those using CC and those using testosterone gels were comparable (525 ng/dL vs 412 ng/dL). Serum estradiol levels were also higher in men receiving testosterone injections, compared to men using CC, those using testosterone gels, and those not receiving TST (6.0 vs 2.0, 2.0, and 2.0 ng/dL, respectively; each P < .01).

Continue to: The qADAM scores...

The qADAM scores for hypogonadal symptoms showed no significant difference in men treated with CC, testosterone injections, and testosterone gels and those not receiving TST (35, 39, 36, and 34, respectively). Men receiving testosterone injections reported greater libido (range, 1-5) than men using CC, those using testosterone gels, and those not on TST (4 vs 3, 3, and 3; P = .047, .04, and < .01, respectively), but it is uncertain if this is clinically meaningful.

Enclomiphene citrate demonstrates improvement in hormone levels

A 2014 Phase II RCT investigated the effects of oral EC—a trans-isomer of CC—compared to topical testosterone 1% gel (T gel) in 124 men with secondary hypogonadism.³ Entry criteria included a baseline morning total testosterone level of < 250 ng/dL on 2 occasions. Participants were divided into 4 groups: 12.5-mg dose of EC, 25-mg dose of EC, T gel, and placebo.

The EC groups and the T gel group had comparable increases in testosterone levels after 3 months of treatment, without statistical significance. The 3-month change in serum testosterone level from baseline was 217 to 471 ng/dL (95% CI, 399-543) in the 12.5-mg dose group; 209 to 405 ng/dL (95% CI, 349-462) in the 25-mg dose group; and 210 to 462 ng/dL (95% CI, 359-565) in the T gel group. The placebo group had a decrease in serum testosterone levels, from 213 to 198 ng/dL (95% CI, 171-226).

Men with low or low-normal serum luteinizing hormone levels may be good candidates for the use of SERMs for management of testosterone deficiency.

Also, the EC groups demonstrated increases in LH and FSH levels from baseline to 3 months, while the T gel group showed a suppression (to low-normal range) in both levels: LH, 1.4 mIU/mL (decrease of 4.4 mIU/mL) and FSH, 2.4 mIU/mL (decrease of 2.4 mIU/mL). Among a subset of men (n = 67) who had at least 2 assessments at the end of 3 months, the researchers also analyzed changes in sperm concentration, using the lower limit of normal (15 million/mL). The number of men with a low sperm concentration increased significantly in the topical T gel group (16% to 53%) compared to the 12.5-mg EC group (decrease from 16% to 12%; P = .0008) and the 25-mg EC group (decrease from 5% to 0%; P = .0007), as well as compared to the placebo group (increase from 8% to 15%; P = .007).

With EC, testosterone remains elevated after treatment cessation

A 2016 2-center parallel, double-blind, ­placebo-controlled RCT evaluated the effect of 2 doses of EC (12.5 mg and 25 mg; n = 85) vs testosterone gel (1.62%; n = 85) vs placebo (n = 86) on serum testosterone, LH, FSH, and sperm counts in 256 overweight and obese men ages 18 to 60 years who had 2 morning testosterone measurements < 300 ng/dL and a low or inappropriately normal LH level for 16 weeks.⁴ All baseline characteristics, including age, BMI, sperm concentration, and serum total testosterone were statistically consistent within groups at both centers. For men receiving EC who did not achieve a testosterone level > 450 ng/dL, there was an up-titration from 12.5 mg to 25 mg at Week 4.

Continue to: All active treatment groups...

All active treatment groups showed increases in testosterone level during treatment (P < .001); however, FSH and LH levels increased in the EC group and decreased in the testosterone gel group (P < .001). Serum testosterone levels improved to 428.8 ng/dL (95% CI, 395-462) and 368.8 ng/dL (95% CI, 307-431), respectively, in the combined EC and testosterone gel groups at 16 weeks. Of note, total testosterone levels after cessation of treatment (off-drug point) rapidly decreased below baseline in the testosterone gel group compared to the pooled EC group, which remained elevated above baseline for at least 7 days.

Composite end-point analysis was performed, with success considered if men achieved both testosterone in normal range (300-1040 ng/dL) and sperm concentrations ≥ 10 × 10⁶. The pooled data studies showed EC was more successful than testosterone gel in achieving both endpoints (63.5% vs 24.7%; P < .001). No difference in the incidence of treatment-related adverse effects between groups was noted.

There were no major adverse effects, even after 3+ years of treatment

A 2019 retrospective cohort study of 400 men treated for symptomatic hypogonadism with CC sought to determine if improvements in testosterone, hypogonadal symptoms, and adverse effects were similar for those treated for ≤ 3 years (n = 280) and those treated for > 3 years (n = 120).⁵ Outcomes included serum testosterone and estradiol levels, symptom improvement (by qADAM questionnaire), and adverse effects.

All participants had a baseline testosterone level < 300 ng/dL, and all participants received CC therapy. Men received 25 mg/d with titration to 50 mg/d when testosterone did not improve to ≥ 300 ng/dL after 4 weeks.

When comparing outcomes across the 2 groups, there were no significant differences. Serum testosterone levels improved to 579 ng/dL (95% CI, 554-605) and 542 ng/dL (95% CI, 504-580) in the ≤ 3 years and > 3 years groups, respectively. Meanwhile, 79% of men in the ≤ 3 years group reported symptom improvement (improvement in libido, erection, or 3 other of the 10 domains of the qADAM questionnaire), while 77% of those in the > 3 years group reported improvement (P = .60).

Continue to: Finally, the percentage of men reporting...

Finally, the percentage of men reporting adverse effects did not significantly differ between groups: 9% in the ≤ 3 years group and 8% in the > 3 years group (P = .85). The most common adverse effects reported in order of frequency were mood changes, blurred vision, breast tenderness, hypertension, hematocrit changes, and flushing. No major adverse events (eg, myocardial infarction, cerebrovascular accident, venous thromboembolism, suicidal behavior) were reported in any patients.

Both the American Urological Association and the Canadian Urological Association support the use of SERMs, especially in hypogonadal men who are interested in fertility preservation.

Of note, although measured estrogen levels at the end of treatment were similar for both groups (54.8 pg/mL in the ≤ 3 years group vs 54.6 pg/mL in the > 3 years group), 37% of patients treated for > 3 years did receive anastrozole treatment for hyperestrogenism compared to 15% in the ≤ 3 years group (P = .05). The authors caution, though, that due to only 20% of the cohort patients having data on pre- and post-treatment estradiol levels, the study was likely underpowered to detect true differences among subgroups.

Recommendations from others

Current American Urological Association and Canadian Urological Association Guidelines note that while greater study on nontraditional testosterone therapies is needed, both organizations support use of SERMs, especially in hypogonadal men who are interested in fertility preservation, as increases in endogenous serum testosterone production do not impact fertility potential, unlike exogenous hormonal replacement.^6,7 Additionally, men with low or low-normal serum LH levels may also be good candidates for the use of SERMs for management of testosterone deficiency.

Editor’s takeaway

Laboratory data (disease oriented) consistently shows that SERMs effectively increase testosterone levels to those comparable with testosterone gels. SERMs resulted in higher semen counts and maintained LH and FSH levels, but there were instances of hyperestrogenism. Data on longer-term benefits and adverse effects of both SERMs and testosterone supplementation are still needed.