Systematic review: The role of bicarbonate therapy after urinary diversion

Abstract

Background

Radical cystectomy (RC) with urinary diversion remains the standard of care for muscle invasive bladder cancer. While metabolic acidosis (MA) after RC is common, the relationship between MA, post-operative complications, and the impact of bicarbonate supplementation remains unclear.

Objective

To review the role of bicarbonate therapy for MA following RC.

Methods

We performed a comprehensive systematic review according to Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines analyzing MA after RC. We included randomized controlled trials and cohort studies that reported metabolic acidosis complications and studies that utilized bicarbonate therapy peri-operatively. Our primary outcome was metabolic acidosis or role of metabolic acidosis as it relates to bicarbonate therapy or post-operative complications.

Results

Twenty-nine studies were identified to investigate MA in the setting of postoperative complication rates or the role of bicarbonate treatment in peri-operative pathways. 19 studies evaluated the incidence of MA without intervention, 5 studies supplemented bicarbonate therapy for patients meeting specific criteria, and 5 studies utilized a standardized postoperative bicarbonate therapy regimen. The incidence of post-operative metabolic acidosis ranged from 1–86% after continent urinary diversion and 1–27% after ileal conduit urinary diversion. Post-operative renal function, diabetes, and older age were factors associated with metabolic acidosis. There was lack of uniformity, indications, or consistent outcomes for utilizing postoperative bicarbonate therapy.

Conclusion

There is little uniformity regarding the role of supplemental bicarbonate therapy and its impact on post-operative complications following urinary diversion. Future studies are needed to better evaluate the role MA plays in peri-operative RC pathways.

Introduction

Radical cystectomy (RC) followed by urinary diversion (UD) remains the standard of care for treatment of muscle-invasive bladder cancer. Given the high morbidity rates following surgery, enhanced recovery after surgery (ERAS) protocols have been studied and implemented to help reduce 30-day readmission rates and complicates. Interestingly, these protocols have demonstrated the leading causes of readmission are UTI and dehydration with another study identifying upper urinary tract obstruction, pyelonephritis, intestinal obstruction, and metabolic acidosis.^1,2 However, metabolic acidosis (MA) has yet to be studied or extensively targeted in a systematic fashion.

Urinary diversions include incontinent diversions, such as ileal or colonic conduit, and continent diversions, such as neobladders or continent cutaneous diversions. MA following RC with UD is thought to arise from repeat or persistent exposure of urine to the bowel mucosa resulting in structural changes due to electrolyte transport favoring ammonium chloride absorption and bicarbonate and potassium secretion. ³ Emphasizing the common nature of electrolyte imbalances postoperatively, MA is seen in 50% of patients one month after neobladder reconstruction, and the severity of MA varies based on factors such as the type of UD performed.^4,5 Although often described as an acute complication, MA may persist even 2 years after UD is performed. ⁶ Specifically, risk factors associated with development of MA include lower eGFR and urinary tract obstruction, independent of age and UD type. ⁷

Despite the well-established nature of MA following RC/UD, there is limited understanding of whether MA is an independent predictor of postoperative complications. Though early postoperative MA may present sub-clinically, chronic MA may increase the risk of bone demineralization and fractures in addition to nephrolithiasis.^8,9 Previous work has studied the relationship of postoperative bicarbonate levels and complications following RC such as risk of readmission for failure to thrive. ¹⁰ These findings have prompted the role of alkaline therapy in treating MA, but there is a lack of standardization or research studies showing the association between correcting MA and decreasing postoperative complications. As such, we sought to establish a comprehensive review of the current role of bicarbonate therapy for MA following RC/UD.

Methods

Search strategy

A comprehensive literature search of PubMed, Embase, SCOPUS, and Cochrane Library utilizing subjects and keywords relating to cystectomies and bicarbonate was performed on 2/10/2022 for studies from 2011-present (Figure 1; see Appendix for search terms). Articles in English that were randomized controlled trials (RCTs) and/or retrospective or prospective cohort studies reporting on metabolic acidosis complications or bicarbonate therapy and/or re-admission outcomes of cystectomy, urinary diversion or reconstruction surgery in relation to electrolyte disturbances were included. We excluded articles that 1) were not in English, 2) were not RCTs or cohort studies, 3) only contained an abstract, or 4) did not have a major focus on metabolic complications or bicarbonate therapy. Our literature review was conducted and reported according to PRISMA-2020 guidelines.

Figure 1.

PRISMA 2020 flow diagram of study identification and inclusion.

Study selection

A total of 1047 citations were initially identified with our search and after deduplication, 650 unique studies met criteria for title/abstract screening in Covidence (Figure 1). Two authors (SD, JF) screened all abstracts identified in the literature search. If eligibility was not agreed upon by initial abstract review, disagreements were resolved by a third reviewer (KG). After exclusions, 92 articles met criteria for full-text extraction. Articles were then reviewed for full-text level inclusion (SD, JF). Reasons for exclusion included: abstract only (34), wrong study design (18), wrong outcomes (9), and not in English (1). Any discrepancies were resolved with secondary review by both authors until a consensus was achieved. Full-text article citations were examined for additional references that may meet inclusion criteria.

Ultimately, our review included 29 retrospective and prospective cohort studies investigating postoperative bicarbonate treatment, metabolic disturbances, and other postoperative complications associated with MA following radical cystectomy and urinary diversion. The studies were then further stratified by those using standardized versus criteria-specific use of bicarbonate supplementation.

Outcomes

Our primary outcomes of interest were metabolic acidosis as a primary endpoint and the presence or absence of bicarbonate therapy in treating electrolyte disturbances following RC/UD. Secondary outcomes included type of urinary diversion and type of study performed. The following study outcomes were extracted: type of study (RCT vs cohort), incidence of MA, the relationship of MA in relation to ileal conduit vs. neobladder or both, and if there was standardization of bicarbonate therapy postoperatively.

Statistical analysis

Due to heterogeneity of studies and limited studies on role of bicarbonate therapy in metabolic acidosis following RC/UD, a meta-analysis was not performed. However, the study data was summarized (Table 1) and evaluated for trends. When statistically significant findings are described in this review, these refer to the findings within individual studies.

Table 1.

Summary of included studies divided by inclusion and type of sodium bicarbonate therapy.

Studies Pertaining to MA Directed Sodium Bicarbonate Therapy
Study Author	Study Type (RCT vs cohort)	Type of Urinary Diversion Performed	Incidence of MA	Risk Factors for MA	Bicarbonate Supplementation Therapy
Chan et al.	Retrospective study	Studer neobladder diversion: 72/72 (100%)	10/72 (13.9%)	Not reported	4.2% (3/72) treated with oral alkaline therapy. Regimen not specified.
Cho et al.	Retrospective study	Ileal neobladder: 62/95 (65%) Ileal conduit: 33/95 (35%)	At one month: Ileal neobladder: 18/58 (31.0%), Ileal conduit: 4/27 (14.8%); At one year: Ileal neobladder: 11/48 (22.9%), Ileal conduit: 2/20 (10%)	Lower GFR and urinary tract obstruction	3/95 (3.1%) at one month; 6/95 (6.3%) at one year; Oral sodium bicarbonate, specific regimen not specified. MA normalized in all patients except for two.
Daneshmand et al.	Prospective study	Orthotopic neobladder: 70/110 (64%); Continent cutaneous diversion: 5/110 (4%); Ileal conduit: 35/110 (32%)	Not reported	Not reported	Oral sodium bicarbonate was prescribed postoperatively if discharge bicarbonate level was less than 22 mmol/L. Dosages not reported.
Hautmann et al.	Retrospective study	Ileal neobladder: 1013/1540 (66%); Remainder not specified.	646/923 (70%) of patients after RC and ileal neobladder; Only 923 patients with follow up > 90 days were included in the analysis.	Not reported	Oral 2–6 gm of sodium bicarbonate stopped within 1 year except for 33% of patients.
Müller, Guido, et al.	Retrospective study	Ileal neobladder: 345/345 (100%)	200/345 (58.0%) at start of rehab; 86.7% at end (3 weeks later) (MA defined as blood pH < 7.35 or base excess of at least −2.5 mmol/l)	Decreased urinary leakage and urine loss less than 1000 gm on the 24-h pad test	Median 3 g (interquartile range 2–5) oral sodium bicarbonate treatment for a base excess of > 2.5 mmol/l
Helmy et al.	Prospective study	Ileocecal neobladder: 19/28 (68%); Ileal neobladder (Studer): 5/28 (18%); Sigmoid neobladder: 4/28 (14%)	4/28 (14.3%; 3 mild, 1 severe; 3 declined supplementation); Ileocecal NB: 2/19 (10.5%); Ileal NB 2/5 (40%), sigmoid NB 0/4 (0%)	Not reported	1 patient with severe MA received IV bicarb, all other patients received oral bicarbonate supplementation following surgery (dosing regimen not reported)
Harraz et al.	Retrospective study	Orthotopic Neobladder :520/895 (58%); Ileal conduit: 348/895 (39%); Other (not specified): 27/895 (3%)	Overall MA incidence not reported. Of 1st readmissions, 20/177 (11.3%) for MA	Not reported	All patients received sodium bicarbonate (dosing not reported) for first 3-months postoperatively
Kim et al.	Retrospective study	Ileal neobladder: 123/123 (100%)	64/123 (52%); At 1 year after surgery, MA incidence decreased 24/123 (19.5%)	1 month post-operatively: renal function; 1 year post-operatively: diabetes; 2 years post-operatively: age and diabetes	All patients were prescribed oral sodium bicarbonate for 3 months (dosing not specified) after urethral catheter removal.
Mahmoud et al.	Prospective study	Ileal neobladder: 200/200 (100%)	3-month sodium bicarbonate group: 32/100 (32%); Chronic sodium bicarbonate group: 21/100 (21%) (No significant difference) (MA defined as venous bicarbonate level of <21 mmol/L)	Not reported	100 patients maintained on 1–2 mEq/kg/day of oral sodium bicarbonate for minimum 1 year vs 100 patients who stopped alkaline therapy after 3 months
Yadav et al.	Prospective study	Ileal neobladder: 42/42 (100%)	3/42 (7.1%) at 1 year (MA (defined as pH < 7.35 or bicarbonate < 22.5 mmol/mL)	Not reported	500 mg oral sodium bicarbonate three times daily for all patients for 6 months post-operatively and continuation if warranted based on ongoing metabolic disturbances.

*This study had discrepancy between percentages of their total patient population. Percentages were calculated based off total number of patients in each specific category.

Results

Metabolic disturbances

We divided the 29 included studies into 3 main categories regarding metabolic acidosis: metabolic acidosis with no supplementation with bicarbonate therapy (19 studies),^6,10–27 MA with bicarbonate treatment for patients meeting study specific criteria (5 studies),^1,7,28–30 and MA with standardized postoperative bicarbonate therapy (5 studies).^2,4,31–33

Type of urinary diversion performed

The most common form of urinary diversion evaluated were neobladders and ileal conduits. Out of the 29 studies, 9 studies evaluated only neobladders, 1 evaluated only IC, 1 study did not specify diversion type, and the remainder evaluated a mix of urinary diversion types. Across all the studies included, the incidence of MA following neobladder ranged from approximately 1 to 86%,^{4,10,17,19,28–33} while the incidence of MA following ileal conduit urinary diversion ranged between 1 to 27%.^{6,7,16,21–23,26} In studies comparing MA between types of urinary diversion, the overall incidence of MA ranged from approximately 1 to 40%.^{6,7,12,16,21,23,26,31} When limiting to studies that reported incidence of MA in both neobladders and ileal conduits, neobladder demonstrated a higher incidence of MA than ileal conduit in 4/5 studies (Table 1).^6,7,21,23,26 Specifically, one retrospective study with 65% ileal neobladders and 35% ileal conduits reported a statistically non-significant difference in the incidence of MA in 31% of patients with an ileal neobladder and 15% of patients with an ileal conduit 1 month postoperatively. ⁷

Factors predictive of metabolic acidosis following RC/UD

Metabolic acidosis or electrolyte abnormalities were reported as postoperative complications in 27 of 29 studies (Table 1). Within individual retrospective studies evaluating patients who underwent neobladder urinary diversions, the following independent risk factors demonstrated a statistically significant association with MA: lower eGFR and impaired renal function, serum creatinine greater than 1.0 mg/dl, urinary tract obstruction, diabetes, and decreased urinary leakage.^4,7,30 Additionally, in a retrospective study of 123 patients who underwent neobladder urinary diversion, Kim et al. identified that at 2 years post-operatively, age and diabetes were both statistically significant risk factors associated with persistent MA. ⁴

Additionally, in a retrospective study of 643 patients, Tyson et al. identified that patients > 75 years old who undergo continent diversions and experience a decrease in serum CO2 levels greater than or equal to 6 mmol/l are at the highest risk of readmission for failure to thrive. ¹⁰ In a separate retrospective study of 95 patients that received either ileal conduit (35%) or ileal neobladder (65%), low eGFR was associated with MA at both 1 month and 1 year post-operatively, while urinary tract obstruction was a risk factor at 1 year. ⁷ Within that study, at 1 month post-operatively, serum creatinine was statistically significant in correlation with the serum bicarbonate and pH level, indicative of MA severity. ⁷

Postoperative complications after RC/UD

Independent risk factors for post-operative complications after RC/UD were evaluated in 16 of 29 studies. These factors studied were not necessarily specific to MA, but multiple studies included bicarbonate levels in the predictive models. Across two individual studies, BMI, incontinent urinary diversions, and Charleson Comorbidity Index ≥ 2 were identified as independent risk factors for complications after RC.^13,14 Independent predictors for early readmission included high grade postoperative complications, orthotopic bladder substitution, and prolonged hospital stay. The most common causes of readmission after RC included upper urinary tract obstruction, pyelonephritis, intestinal obstruction, metabolic acidosis, UTI, and dehydration.^1,2 Additionally, in a retrospective study, evaluation of 1 week postoperative lab values following RC such as median daily white blood cell count, bicarbonate, and creatinine increased the ability for models to identify patients at highest risk for readmission within 30 days of RC/UD. ¹⁷ For example, readmitted patients had significantly greater variance in white blood cell counts than in non-readmitted patients. ¹⁷

Bicarbonate therapy and independent risk factors for MA

Of 29 studies, 34% (10/29) discussed the role of bicarbonate therapy in addressing MA following RC/UD. Of these studies, 5 prescribed bicarbonate therapy for patients meeting criteria for MA specific to the patient or study^1,7,28–30 while 5 studies prescribed standardized post-operative bicarbonate therapy prophylactically to all patients.^2,4,31–33

Non-universal bicarbonate therapy

Of the 5 studies that prescribed bicarbonate therapy based on specific criteria, 2 studies evaluated MA following ileal neobladder. For patients in which MA was defined as a pH of less than 7.35 or a base excess of greater than 2.5 mmol/l, 86% were prescribed bicarbonate supplementation based on these criteria. ³⁰ Based on this data, Muller et al., recommend sodium bicarbonate treatment for a base excess of > 2.5 mmol/l and to closely monitor patients with renal insufficiency for MA. ³⁰ In a study of 72 patients undergoing neobladder urinary diversion, Chan et al., reported electrolyte abnormalities in 13.9% (10/72) of patients and non-standardized treatment with alkaline therapy in 4.2% (3/72). ²⁸ In that individual study, there was no association between electrolyte disorders or bicarbonate therapy with urinary tract calculi. ²⁸

The remaining 3 studies reporting non-universal bicarbonate supplementation included multiple types of urinary diversions. Cho et al. compared MA in ileal neobladders vs conduits and defined MA as a bicarbonate level lower than 21 mmol/L. ⁷ Nine patients with neobladders were treated with sodium bicarbonate for bicarbonate < 15 mmol/L in the setting of renal insufficiency with “most of these patients restoring a normal metabolic status” and none requiring hemodialysis. ⁷ Six patients remained on bicarbonate supplementation at one year post-operatively. ⁷ Similarly, in a retrospective study, Hautmann et al. reported MA in 70% of 1540 patients after RC and ileal neobladder with 33% of patients being treated with oral 2–6 gm of sodium bicarbonate daily for over 1 year. ²⁹ Prior to starting supplementation, 11/923 (1.2%) patients were readmitted due to a severe “salt losing, hypovolemic and acidotic state, including 5 with impaired renal function and 5 in retention.” ²⁹ Finally, Daneshmand et al. prescribed oral sodium bicarbonate postoperatively if discharge bicarbonate level were less than 22 mmol/L. ¹

Universal standardized bicarbonate therapy

Of the 5 studies with universal bicarbonate therapy, 4 focused on patients with neobladder urinary diversions and 1 included multiple types of UD.^2,4,31–33 In a retrospective study of 123 patients who underwent neobladder urinary diversion and were prescribed oral sodium bicarbonate supplementation (dosing not specified) for 3 months after urethral catheter removal, 52% of study participants developed MA during the first month post-operatively. ⁴ At 2 years, MA incidence decreased to 12.1% of patients. ⁴ Similarly, in a prospective study of 42 patients who underwent ileal neobladder and were treated prophylactically with 500 mg of sodium bicarbonate three times daily for 6 months, Yadav et al. reported that 3/42 (7%) patients met MA criteria (pH < 7.35 or bicarbonate < 22.5 mmol/mL) to continue therapy at the 1 year mark. ³³ No patients in this study reported symptomatic MA. ³³

In patients with a solitary kidney who underwent orthotopic neobladder, 14.3% of patients developed MA despite prescription of universal post-operative oral sodium bicarbonate therapy. ³¹ However, of the 4 patients who developed MA, 3 patients with mild MA (base excess > −5 Umol/L) did not take the sodium bicarbonate therapy. One patient developed severe MA (base-excess < −5 Umol/L) and treatment consisted of readmission, urinary catheterization, and IV bicarbonate therapy. ³¹ In a prospective separate study comparing chronic alkali therapy > 1 year (1–2 mEq /kg/day of sodium bicarbonate) to therapy for 3 months (self-discontinued above dosing due to side effects) in 200 patients with ileal neobladders, there was no overall difference in serum bicarbonate levels or blood pH after 1 year and no patients were readmitted for severe MA. ³² This study had 80% power to detect a 30% difference in the incidence of metabolic acidosis between the two study groups. ³² Finally, in a retrospective study of 895 patients undergoing either orthotopic neobladder, ileal conduit, or UD not specified, metabolic acidosis remained a frequent cause for admission (11.3% of readmissions) despite all patients receiving sodium bicarbonate supplementation for 3 months post-operatively. ² MA was also one of the most common readmission causes (13.6%) for patients with three or more readmissions. ²

Discussion

Metabolic acidosis is a common occurrence in the postoperative setting of cystectomy with urinary diversions, occurring in up to 70% of patients. Although the mechanism for these electrolyte abnormalities has been extensively documented in the literature, there has been very little research to determine the efficacy of preventing postoperative metabolic acidosis, and whether metabolic acidosis accurately predicts postoperative complications. Furthermore, there is a lack of consensus on the optimal clinical management of metabolic acidosis after urinary diversion. This review investigated previously reported independent risk factors for metabolic acidosis and the effect of MA as a predictor for postoperative complications following RC/UD. We collectively grouped prior studies to elicit key conclusions that may allow further investigation of early identification and targeting of metabolic acidosis to help predict and prevent complication rates in specific patient populations.

Five papers meeting inclusion criteria evaluated independent risk factors for metabolic acidosis. Unfortunately, there was little uniformity between these studies, making it difficult to draw generalized conclusions. For example, the definition of metabolic acidosis was variable, with some utilizing absolute bicarbonate values, whereas others used relative base excess or blood pH.^3,7,10,29,32 Despite these limitations, postoperative metabolic acidosis correlated with impaired post-operative renal function and diabetes.^4,7 Metabolic acidosis and CKD may act synergistically to promote poor outcomes, as research shows that an acidotic state in the setting of CKD can stimulate promotion of prone inflammatory and fibrotic states with subsequent worsening of kidney function and further exacerbation of an acidotic state. ³⁴ By more carefully observing this specific population of patients, we may identify patients earlier that are at higher risk of post-operative complications or consider prophylactically giving bicarbonate therapy to help prevent future complications. Similarly, patients with diabetes are another population that may benefit from closer monitoring for post-operative metabolic acidosis. It has been hypothesized that this correlation may be due to the increased liver lysis and ketogenesis that occurs in an insulin deficient state, which is often present in the postoperative period due to the stress sustained during surgery and anesthesia. ³⁵

Additionally, despite the commonplace nature of both RC in urological surgery and the incidence of MA post-operatively, the role of bicarbonate therapy in treating electrolyte disturbances has yet to be fully elucidated. In this review, 10/29 studies described treating MA following RC/UD with bicarbonate therapy. However, only half reported universal bicarbonate prophylaxis therapy with varying dosing regimens (2–6 g daily) and postoperative monitoring.^2,4,31–33 However, these studies did identify patient factors associated with lower bicarbonate levels, such as female gender and age <65. ⁴ A subset of studies suggested decreased readmission rates when patients received universal bicarbonate therapy,^32,33 while others reported readmission for MA despite therapy.^2,31 Overall, the variability across these studies in treatment regimens and outcome reporting limits the ability to draw conclusions on the impact of bicarbonate supplementation on MA severity and complication rate. This review highlights the existing knowledge gap regarding the role of standardized bicarbonate therapy, either in a prophylactic or clinically directed pathway. A future randomized controlled trial could clarify the impact of universal bicarbonate therapy on post-operative metabolic acidosis and complications following radical cystectomy and urinary diversion.

Conclusion

Metabolic acidosis is a common postoperative complication of radical cystectomy. Although the physiology for post-operative MA is well documented, there is a paucity of literature investigating the role that MA plays in post-operative complications, or if preventing MA can help decrease complications and readmissions. Similarly, although bicarbonate therapy supplementation has been loosely discussed in the literature, the optimal regimen and its overall impact on complication rates remains to be determined. Further studies are needed to more closely evaluate the role of bicarbonate therapy as it relates to post-operative metabolic acidosis, and whether bicarbonate treatment to alleviate MA can help improve perioperative outcomes.