Cost effectiveness in the transition from open to percutaneous nephrolithotomy in a low and middle-income country public hospital setting

There are approximately 24 million incident cases of urolithiasis in low- and middle-income countries (LMICs) each year, representing a significant global health burden [1]. Kidney stones, the most common type of urolithiasis, can cause intense flank pain, hematuria, and recurrent urinary tract infections. If left untreated, they may obstruct urinary flow and contribute to complications such as hydronephrosis, infection, chronic kidney disease and renal failure [1].

Surgical management is often necessary for complex kidney stone disease (KSD), with treatment options ranging from open surgery to various minimally invasive techniques. Historically, open pyelolithotomy, requiring a large abdominal or flank incision to access the renal collecting system, was the standard approach for large or complex stones. However, in most developed countries, percutaneous nephrolithotomy (PCNL) has largely replaced open surgery as the preferred intervention for complex cases. PCNL involves accessing the renal pelvis through a small percutaneous tract, minimizing tissue disruption. Extensive evidence supports PCNL’s advantages over open surgery, demonstrating benefits such as reduced morbidity, shorter hospital stays, faster recovery, and improved cosmetic outcomes compared to open pyelolithotomy [2-5].

In some LMICs, open pyelolithotomy remains the predominant treatment modality, largely due to constraints including limited funds, the capital cost of specialized equipment, lack of training and the need for expertise and different types of team members in minimally invasive surgery. However, literature from high income countries has demonstrated that minimally invasive procedures may be cost-saving in the long-term due to factors such as reduced hospital stay and lower complication rates [2-5].

The objective of this study is to analyze and compare surgical techniques that can be used to treat large kidney stones in the pelvis, namely open pyelolithotomy and PCNL, in an LMIC setting, to assess the potential change in costs of replacing open pyelolithotomy cases with PCNL. We believe that PCNL represents a cost-saving and beneficial alternative for treating KSD in LMICs.

Study Approval and Informed Consent
The study was approved by the Institutional Review Board at University of California, San Francisco (UCSF) (IRB # 21-33758).

Data Collection
We examined case logs of urology patients at Georgetown Public Hospital Corporation (GPHC) in Guyana from January 2018 to December 2024. GPHC introduced PCNL in 2023, thus the PCNL cases were from January 2024 to December 2024. The case logs provided comprehensive data on patients who underwent urologic procedures including age, year, pre-operative diagnosis, procedure, time started, and time finished. Case logs were de-identified, and this data was retrospectively reviewed. Data on hospitalization days, complications and transfusions were gathered via a chart review and this data was used in the extrapolated cost model. The associated costs of these parameters were analyzed to assess the economic impact of different treatment approaches for KSD.

Patient Selection
From the de-identified case logs, patients were stratified based on the type of stone (upper versus lower urinary tract) and the type of procedure performed. The procedures were categorized as open or minimally invasive. Open procedures to treat urinary stone disease at GPHC included open nephrectomy, nephrolithotomy, pyelolithotomy, and ureterolithotomy. For the purposes of this cost comparison analysis, we studied costs associated with pyelolithotomy cases.

Cost parameters and cost effectiveness analysis
A retrospective chart review of the 59 patients who received open pyelolithotomy and 20 patients who received PCNL was conducted to obtain data on days of hospitalization, complications and transfusions. Due to lack of granular data on care provider fees and per minute operating room fees, as these data were not available on a per unit cost in this public hospital setting, comparable published standardized costs were extrapolated in our model.[6] Post-surgical hospital stay cost was based on data scaled by respective US and Guyana gross domestic product (GDP) per capita.[6-9]  The decision tree analysis compared the costs and probabilities of each parameter to obtain an overall average cost.

Preliminary Results
A retrospective review of GPHC urology patient case logs from January 2024 to December 2024 was conducted. Pyelolithotomy was the most common open stone procedure. Within the identified open cases, 28 cases with complete data were used in the final decision tree analysis. The average age of patients who underwent open pyelolithotomy was 48.4 years old. Pre-operative diagnosis for patients undergoing open pyelolithotomy included renal, lower calyx, ureteropelvic junction (UPJ) stones and staghorn stones. The mean operative time for a patient undergoing pyelolithotomy was 1 hour 38 minutes and the average length of hospitalization was 7.0 days. The average age of patients who received PCNL was 53.4 years old. Preoperative diagnosis included renal pelvis stones, upper calyx, middle caylx, and lower calyx stones. The mean operative time was 2 hours 1 minute, and the average length of hospitalization was 3.7 days. The Wilcoxon rank-sum test was used to compare continuous variables between the two groups and found no significant difference in patient age (P = 0.2), but significant differences in operative time and length of hospitalization (P = 0.006, P = 0.0001, respectively) (table 1).

Decision Tree Results
Quantitative data was analyzed using a decision tree to show the extrapolated value, or weighted average cost, of open pyelolithotomy versus PCNL. Table 2 and 3 give the costs for the various branches of the decision tree. Complications included bleeding, infection and fever as documented in the case logs and patient charts. The decision tree classifies all transfusions along the complication path, therefore the total percentage of transfusions across all patients is 50.0% * 25.0% = 12.5% for open surgery and 35.0% * 28.6% = 10.0% for PCNL. The Fisher’s exact test was used to assess associations between binary variables and found no significant differences in the probability of complication (P = 0.35) or probability of transfusion (P = 0.28) between the two groups (table 3).

For the decision tree analysis, the total surgery cost ($3,414 for open pyelolithotomy and $2,449 for PCNL) is the average days in hospital multiplied by the hospital cost per day plus operation cost (including operating room, surgeon, and anesthesiologist costs), as outlined above in table 2 and 3. Costs from the literature have been scaled to Guyana costs using GDP data. For example, if a cost was obtained for a US procedure from a paper written in 2017, then the equivalent Guyana cost in 2024 is assumed to be: US cost multiplied by (Guyana 2024 GDP per capita / US 2017 GDP per capita).

Figure 1 shows the decision tree with the expected outcome. The complication cost was calculated by the cost per day multiplied by the extra hospital days. Each path on the tree represents a possible scenario with associated probabilities and costs. Starting from the decision node (yellow box), the tree branches into open pyelolithotomy and PCNL, each with their own set of probabilities for potential complications and outcomes. Open pyelolithotomy was assigned a 50.0% chance of complications and a 50.0% chance of no complications (based on GPHC primary data), with further branching based on the need for a transfusion. PCNL was assigned a 35.0% complication rate and a 65.0% chance of no complications (based on GPHC primary data), also with further branching based on the probability of transfusion. Each endpoint of the branches provides the expected costs and outcomes, allowing a comparison of overall cost-efficiency between the two surgical options. The left boxes present the cost on the forward path, and the right boxes present the costs on the backward path (considering all downward pathways). This analysis suggests open pyelolithotomy is approximately 40% more than the cost of PCNL ($4160 versus $2974). The primary drivers of the higher cost of open pyelolithotomy compared to PCNL were longer hospitalization days and higher complication rates, as complications were assigned a cost per day.

Kidney stone disease poses a significant burden on global healthcare, necessitating effective and cost-efficient treatment strategies. In high income countries, open surgical procedures to treat kidney stones are rarely indicated because literature from high income countries has demonstrated that minimally invasive procedure alternatives such as PCNL have better health outcomes and lower costs [11]. However, there is limited research on the financial burden of treating kidney stones in LMICs.

Two approaches to KSD treatment, open pyelolithotomy and PCNL, were evaluated in this study via a decision tree analysis modeling the hypothetical cost reduction of entirely replacing open pyelolithotomy with PCNL at GPHC in Guyana. Based on a model using hospital length of stay, complication rates, surgical costs and transfusion rates, we found the average cost of open pyelolithotomy is 40% more than PCNL. A sensitivity analysis of the decision tree results shows the biggest driver of the difference in cost between open pyelolithotomy and PCNL is the length of stay in hospital (7.0 days for open versus 3.7 days for PCNL, P = 0.0001). By reducing hospitalization periods, PCNL offers a promising avenue for cost reduction. This financial benefit supports the argument for transition to minimally invasive techniques like PCNL in LMICs.

Within the past five years, Guyana has transitioned from a lower middle- income country to an upper middle-income country based on the World Bank Classification due to rising gross domestic product from discovery and development of offshore oil reserves [12]. This data captures this transition period. Despite this economic growth, income inequality and healthcare infrastructure gaps persist [12-14]. GPHC serves as the primary public hospital in Guyana and remains the sole provider of surgical subspecialty care in the region for individuals who lack access to private healthcare. The high prevalence of open surgery cases in many LMICs relative to high income countries where open stone surgery is very rare identifies an opportunity for optimizing resource utilization and cost-savings in many LMIC hospitals [15-17]. By replacing a significant proportion of open pyelolithotomy cases with PCNL, substantial cost savings can be achieved while reducing patient morbidity and convalescence. The findings of this study support the transition to PCNL for the treatment of complex kidney stone disease in hospital settings where open stone surgery is common.

By highlighting the economic advantages of PCNL, we aim to encourage healthcare providers and policymakers to allocate the necessary resources and training to expand the utilization of this technique. The implementation of PCNL in LMIC settings holds the potential to improve patient outcomes, reduce costs, and enhance healthcare efficiency [4,7,18,19]. This research demonstrates that an investment up to $1186 per patient to transition from open pyelolithotomy to PCNL is potentially cost saving based on expenses reduced due to averted complications, hospitalization days, and transfusion rates. Although not analyzed in this paper, required equipment for PCNL (nephroscope and digital equipment) costs around $9000 [20]. Two sets of this equipment would cost around $18,000 (doubled as these instruments can have a shorter lifespan than instruments for open surgery). In situations where imaging equipment is not already available, the cost of a compact C arm, which is used for puncture at the start of a PCNL procedure, is around $70,000 [21] and the cost of an ultrasound machine can be considerably cheaper, around $17,000 [22]. Given the cost savings between PCNL and open pyelolithotomy, it would take approximately 30 to 74 patients (depending on method used) to pay for this required equipment. Additionally, reprocessing flexible endoscopes has been estimated to cost between $100 to $300 and require around 1.5 hours of hands-on staff time, which is also important to consider in the upfront investment [23].

It is important to acknowledge the limitations of this study. Most significantly, the study was limited as length of hospitalization, transfusion rates and complication data were only available for a subset of the open pyelolithotomy and PCNL cases due to incomplete charting at GPHC. Additionally, the economic analysis relied on published and scaled data to extrapolate costs due to lack of granular data on care provider fees and per minute operating room fees at GPHC. Translation of costs from a Western country to a LMIC based on relative GDP may not be fully accurate as hospital stay costs are typically much higher in the West. This study was also limited by the dearth of recent literature on open surgeries to treat KSD as these methods are now rarely utilized in developed countries. Finally, this study does not incorporate the rate of delayed hemorrhage after PCNL (such as arteriovenous fistula or pseudoaneurysm) into the cost effectiveness analysis, given the rarity of such events at 1% or less [24,25]. When these events, while rare, do occur, they can increase the costs of hospitalization and transfusion. Overall, further research is needed to account for infrastructure limitations within the healthcare system when assessing the cost of minimally invasive surgery in LMICs. These limitations may influence the cost effectiveness of upfront expenses associated with acquiring and maintaining specialized equipment for PCNL. Maintenance costs, periodic upgrades, and the investment in educating personnel in utilization and support for constantly evolving technology should also be considered in economic analyses.

While these limitations warrant consideration, they are not uncommon in research using primary data from LMICs, where resource constraints often lead to incomplete charting and limited cost transparency. These limitations should not exclude the inclusion of LMIC-based studies in the broader urological literature, but rather highlight the need for more context-specific research that reflects real-world conditions. Despite data gaps, our analysis offers valuable insight into the feasibility and potential advantages of implementing PCNL in these settings, helping to inform both clinical and policy decisions in resource-limited environments. Continued inclusion of LMIC data is essential to develop equitable, evidence-based urological care worldwide.

In conclusion, our study demonstrates that PCNL may offer a cost saving option for the treatment of kidney stones in LMICs. By implementing this technique, healthcare systems in LMICs can improve patient outcomes, reduce costs, and optimize resource utilization. Future efforts should focus on expanding the infrastructure necessary to make PCNL and other minimally invasive treatments of kidney stones more readily available in LMICs given its potential as a cost-savings intervention.

We would like to express our sincere gratitude to our partner institution, Georgetown Public Hospital Corporation and Dr Rajendra Sukhraj and Dr Ravi Purohit for making this study possible; and to Ava Yap for her mentorship and support. We would also like to thank the UCSF Funding Office and the Harold Varmus Global Health Scholars fund for their financial support.