Emphysematous pyelonephritis (EPN) is an uncommon, fulminant infection of the renal parenchyma characterized by gas formation within the collecting system, renal tissue, or perinephric spaces. It most often occurs in patients with uncontrolled diabetes mellitus and, if unrecognized, carries a high risk of mortality¹. Women are disproportionately affected, with reported female-to-male ratios of approximately 6:1 and a mean age of presentation near 55 years². Before the advent of modern imaging and antibiotics, mortality rates reached nearly 80%³. With advances in computed tomography (CT), broad-spectrum antimicrobials, and minimally invasive drainage, contemporary mortality has fallen to between 13% and 25%⁴. The pathophysiology involves gas-forming organisms—most commonly Escherichia coli and Klebsiella pneumoniae—acting in the setting of impaired host defenses and tissue hypoxia. Diabetes mellitus is present in up to 95% of cases, making it the strongest single risk factor⁵. Other contributors include urinary tract obstruction, nephrolithiasis, and states of immunosuppression⁶. CT has become the cornerstone of diagnosis, not only detecting intrarenal gas with near-perfect sensitivity but also delineating disease extent. The Huang and Tseng classification, derived from CT appearances, remains widely used to stage severity and correlate radiological findings with prognosis⁷. Management strategies have evolved in parallel. Emergency nephrectomy, once considered standard, has largely been supplanted by less invasive approaches. Broad-spectrum antibiotics combined with percutaneous drainage (PCD) have demonstrated high success rates in appropriately selected patients⁸. More recent systematic reviews confirm that stepwise, kidney-preserving strategies are associated with improved survival and reduced morbidity, representing a paradigm shift in practice⁹. Despite these advances, EPN continues to challenge clinicians because of its heterogeneity in presentation, variable response to therapy, and significant risk of long-term renal impairment. Prognosis is increasingly thought to depend not only on radiological staging but also on systemic markers such as hypoalbuminemia, thrombocytopenia, and renal dysfunction at presentation. Identifying which patients will benefit most from conservative or minimally invasive approaches remains an area of ongoing clinical debate. Aim and Objectives This study was designed to reassess prognostic factors in EPN and to evaluate outcomes with conservative and minimally invasive therapies in a contemporary cohort. The primary aim was to analyze predictors of outcome and long-term renal function following non-radical management. Secondary objectives included describing the epidemiology, clinical spectrum, imaging findings, microbiology, and mortality associated with EPN in our setting.

Study Population and Design This prospective observational study was conducted at Government General Hospital, Guntur, attached to Guntur Medical College, from September 2023 to July 2025. All patients admitted with a diagnosis of emphysematous pyelonephritis (EPN) during the study period were included. Inclusion and Exclusion Criteria Inclusion criteria were patients presenting with clinical features of acute pyelonephritis (fever, loin pain, vomiting, altered sensorium, chills) and evidence of gas in the renal parenchyma or perinephric/pararenal tissues on non-contrast CT (NCCT KUB). Exclusion criteria were patients with acute pyelonephritis without intrarenal gas, those with recent urological interventions or catheterization, and patients with recent genitourinary trauma. Sample Size Calculation Sample size calculation was done based on the prevalence rate and margin of error in a recent study- Emphysematous pyelonephritis: An experience with 26 cases. Rays Ahmad Misgar, Idrees Mubarik, Indian J Endocrinol Metab. 2016 Jul-Aug; 20(4): 475–480.11 n = Z2 P (1-P) d2 where n is the sample size Z is the statistic corresponding to level of confidence, P is expected prevalence (obtained from the above study), and d is precision (corresponding to effect size). In our study design, we found n = 42 Data Collection and Variables Clinical and demographic details were documented, including age, sex, diabetic status, presence of obstruction, and comorbidities. Laboratory investigations included serum creatinine, HbA1c, serum albumin, and platelet count. Urine cultures were processed for microbiological confirmation. Risk factors (diabetes, stone disease, obstructive uropathy) and prognostic factors (altered sensorium, dialysis requirement, thrombocytopenia, hypoalbuminemia, deranged HbA1c, CT severity, and polymicrobial infection) were systematically recorded. Imaging and Classification Diagnosis was confirmed using NCCT KUB, which also enabled classification according to the Huang and Tseng system (Figure 1) : ● Class I: Gas confined to collecting system ● Class II: Gas in parenchyma without extrarenal spread ● Class IIIA: Extension into perinephric space ● Class IIIB: Extension into pararenal space ● Class IV: Bilateral disease or solitary kidney involvement Treatment Protocol All patients received broad-spectrum intravenous antibiotics (initially piperacillin-tazobactam or cefoperazone-sulbactam, later modified per culture). Resuscitation and glycemic control were instituted. Management was stepwise: ● Conservative therapy: Antibiotics and supportive care. ● Minimally invasive therapy: DJ stenting, percutaneous nephrostomy (PCN), or percutaneous drainage (PCD). ● Extensive therapy: Open drainage or nephrectomy. Indications for minimally invasive procedures included obstruction, sepsis, or significant renal/perirenal collections (Class II–III). Extensive surgery was reserved for patients with persistent sepsis or collections unresponsive to drainage. Outcome Assessment Patients were followed until discharge and again at six months. Outcomes were categorized as: ● Renal salvage with preserved function ● Renal salvage with impairment ● Chronic dialysis dependence ● Nephrectomy survivor ● Death For statistical analysis, outcomes were also dichotomized into: ● Good outcome: Renal salvage with preserved function ● Poor outcome: All other categories Long-term renal function was assessed by serum creatinine and eGFR (CKD-EPI). DTPA renogram was performed when feasible. Definitions of Prognostic Factors ● Thrombocytopenia: Platelets <100,000/cumm ● Hypoalbuminemia: Serum albumin <3.5 g/dL ● Altered sensorium: Confusion, delirium, stupor, or coma ● Dialysis requirement: Indication for renal replacement therapy ● Deranged HbA1c: Elevated HbA1c reflecting poor glycemic control ● Polymicrobial infection: >1 organism on urine culture Statistical Analysis Data were analyzed using SPSS v20. Continuous variables were compared using the Student’s t-test, and categorical variables with the Chi-square test. p value <0.05 was considered as significant. Univariate analysis was applied to determine associations between prognostic factors and outcomes. Given the modest sample size, multivariate analysis was not performed to avoid unstable estimates. Ethical Approval: The study was conducted after obtaining clearance from the Institutional Ethics Committee (Ref No. GMC/IEC/URO/2023/15). Written informed consent was obtained from all participants.

Study Cohort A total of 42 patients with emphysematous pyelonephritis (EPN) were included over the 2-year study period. Demographics Most patients (76.2%) were between 40–60 years, with only 4.8% under 40 years and 19.0% over 60 years. Female predominance was observed (83.3% vs. 16.7%), consistent with the recognized gender bias in EPN. Clinical Presentation Fever was the most frequent symptom (90.5%), followed by flank pain (76.2%) and nausea/vomiting (71.4%). Fatigue was present in 52.4%, and dysuria in 33.3%. Severe systemic features included altered sensorium (23.8%) and oliguria (19.0%), both strongly linked to adverse outcomes. Rare signs such as gas crepitus and pneumaturia (2.4% each) were late manifestations. Risk Factors Diabetes mellitus was almost universal (95.2%). Stone disease (28.6%) and obstructive uropathy (38.1%) were also common, though these baseline factors did not significantly influence prognosis. Microbiology Escherichia coli was the predominant isolate (54.8%), followed by Klebsiella (7.1%). Polymicrobial cultures (16.7%) and sterile samples (16.7%) were also frequent, the latter likely reflecting prior antibiotic exposure. Radiology According to the Huang–Tseng classification, Class II was most common (38.1%), followed by Class IIIA (23.8%). Advanced disease (Classes IIIB–IV) accounted for 26.2%, while Class I was relatively rare (11.9%). Management All patients received intravenous antibiotics. ● Nearly half (45.2%) were managed with DJ stenting alone. ● A further 28.6% required DJ stent plus PCN/PCD. ● Open drainage (9.5%) and nephrectomy (14.3%) were performed in refractory or advanced cases. ● Guided aspiration was rarely used (2.4%). Treatment escalation correlated with CT severity: stenting alone sufficed in low-grade disease, whereas higher classes often required PCN/PCD or, in refractory cases, radical surgery. (Table 1, Figure 2) CT Class DJ Stenting alone DJ Stenting + PCN/PCD DJ Stenting + Open Drainage DJ Stenting + Nephrectomy Guided Aspiration Class I 4 (80.0%) 1 (20.0%) 0 0 0 Class II 15 (93.8%) 1 (6.2%) 0 0 0 Class III A 0 (0%) 9 (90.0%) 0 0 1 (10.0%) Class III B 0 (0%) 0 (0%) 3 (42.9%) 4 (57.1%) 0 Class IV 0 (0%) 1 (25.0%) 1 (25.0%) 2 (50.0%) 0 Table 1: CT Classification vs Treatment Modalities (n = 42) Outcomes At six months, 73.8% achieved renal salvage with preserved function. Nephrectomy survivors comprised 14.3%, while mortality (4.8%), renal impairment (4.8%), and chronic dialysis dependence (2.4%) were uncommon (Figure 3). For analysis, outcomes were dichotomized: ● Good outcome (renal salvage with preserved function) → 73.8% ● Poor outcome (all other categories) → 26.2% Treatment–Outcome Correlation ● DJ stenting alone had 100% good outcomes, but was confined to low-grade disease. ● DJ stenting + PCN/PCD showed 91.7% success, even in higher-grade disease. ● Open drainage and nephrectomy were associated with poor outcomes, reflecting advanced disease at baseline. Prognostic Factors Univariate analysis showed that altered sensorium (p = 0.001), dialysis requirement (p = 0.009), hypoalbuminemia (p = 0.018), thrombocytopenia (p = 0.018), and higher CT class (p = 0.017) significantly predicted poor outcomes. Odds ratios demonstrated the strongest association for altered sensorium (OR ~13.6), followed by dialysis (OR ~8.0), hypoalbuminemia (OR ~7.2), and thrombocytopenia (OR ~5.2). HbA1c derangement and polymicrobial infection did not achieve statistical significance (Table 2, Figure 4). Prognostic Factor Present/ Absent Good Outcomes (n, %) Poor Outcomes (n, %) χ² value p- value Odds Ratio 95% CI Remarks High CT Grade ≥3A YES 11 (52.4%) 10 (47.6%) 5.72 0.017 4.85 1.05 –22.3 Significant NO 18 (85.7%) 3 (14.3%) Hypoalbuminemia YES 5 (50.0%) 5 (50.0%) 5.64 0.018 7.22 1.59 – 32.7 Significant NO 26 (81.2%) 6 (18.8%) Thrombocytopenia YES 5 (50.0%) 5 (50.0%) 5.64 0.018 5.17 1.14 – 23.4 Significant NO 26 (81.2%) 6 (18.8%) Deranged HbA1c YES 9 (64.3%) 5 (35.7%) 1.86 0.172 3.86 0.95 – 15.7 Not Significant (trend) NO 22 (78.6%) 6 (21.4%) Altered Sensorium YES 3 (30.0%) 7 (70.0%) 10.23 0.001 13.57 2.71 – 68.0 Highly Significant NO 28 (87.5%) 4 (12.5%) Need for Dialysis YES 5 (41.7%) 7 (58.3%) 6.80 0.009 8.03 1.82 – 35.5 Significant NO 26 (86.7%) 4 (13.3%) Polymicrobial Culture YES 5 (71.4%) 2 (28.6%) 0.00 1.000 1.27 0.24 – 6.73 Not Significant NO 26 (74.3%) 9 (25.7%) Table 2: Prognostic Factors vs Outcomes (Univariate Analysis, n = 42)

Demographics The demographic profile of our cohort closely mirrors previous reports. Most patients were between 40–60 years of age, consistent with Huang et al.10 who reported that nearly three-quarters of EPN cases occur in this age group. The female predominance (83.3%) observed in our series also agrees with Reddy and Rao,11 as well as other Indian studies, reflecting the higher prevalence of urinary tract infections and anatomical susceptibility in women. Clinical Spectrum Fever, flank pain, and vomiting constituted the dominant presenting symptoms, echoing earlier descriptions of the classical triad of acute pyelonephritis.12 Systemic manifestations such as altered sensorium and oliguria, though less frequent, were strongly predictive of poor outcomes in our cohort. Similar findings have been noted by Huang and Tseng and later Sokhal et al.,13 highlighting the prognostic weight of neurological compromise and renal shutdown. Rare features such as crepitus and pneumaturia were recorded in only a few cases, confirming their limited diagnostic value. Risk Factors Diabetes mellitus was almost universal in our series (95.2%), in line with Misgar et al. and Huang et al.,10 reaffirming its role as the cornerstone predisposing factor. Obstruction due to stones or other causes was observed in one-third, consistent with prior reports that obstruction and diabetes act synergistically. Interestingly, these baseline risk factors did not significantly influence outcomes in our study, likely because early drainage and antibiotics blunted their effect. Similar conclusions have been reported previously, where systemic severity rather than predisposing factors dictated prognosis.12 Microbiology Escherichia coli remained the leading pathogen (54.8%), with Klebsiella and occasional isolates such as Pseudomonas and Proteus comprising the rest. Polymicrobial infections (16.7%) were not uncommon and, as Sokhal et al.13 have noted, often accompany more severe disease. The relatively high proportion of sterile cultures (16.7%) likely reflects prior antibiotic exposure, underscoring the importance of integrating clinical and radiological findings with microbiology. Radiology and Management Correlation CT-based classification revealed Class II as the most common stage, while advanced disease (III B and IV) accounted for just over a quarter of patients. This distribution is consistent with other Indian cohorts, where delayed presentation often precludes identification of early Class I disease. Management was clearly stage-dependent. Low-grade disease (Class I–II) responded well to DJ stenting alone, whereas higher classes frequently required PCN/PCD. Nephrectomy was reserved for refractory or extensively destroyed kidneys, accounting for 14.3% of cases. These trends reflect the global shift toward renal-preserving strategies first popularized after the Huang–Tseng classification and subsequently reinforced by Sokhal et al.13 Outcomes Renal salvage with preserved function was achieved in 73.8% of patients, while mortality (4.8%) and dialysis dependence (2.4%) were low. These outcomes compare favorably with older reports where nephrectomy and death were common. For instance, Huang and Tseng reported mortality rates near 25% in 2000, whereas modern cohorts, including Sokhal et al. (2017),13 describe nephrectomy rates around 20% with mortality under 15%. Our findings reinforce this paradigm shift: with prompt recognition and minimally invasive drainage, most patients can avoid radical surgery and preserve renal function. Prognostic Factors Several systemic markers emerged as significant predictors of poor outcome. Altered sensorium carried the highest risk (OR ~13.6), followed by dialysis requirement (OR ~8.0), hypoalbuminemia (OR ~7.2), and thrombocytopenia (OR ~5.2). These align closely with previous literature, where sensorium changes, sepsis-induced thrombocytopenia, and hypoalbuminemia have been consistently linked to mortality.10,13 Misgar et al. also emphasized hypoalbuminemia and thrombocytopenia as independent prognosticators. While higher CT grade (≥III A) was associated with worse outcomes, its prognostic value was less robust than systemic factors—likely reflecting improved outcomes even in advanced radiological disease when managed aggressively with minimally invasive drainage. Deranged HbA1c, though biologically plausible, did not achieve statistical significance in our cohort. This may be because once infection is established, acute systemic factors outweigh the influence of chronic glycemic control, as also noted in other contemporary series. Risk Factors vs Outcomes Baseline risk factors such as diabetes, stones, and obstruction were not independently predictive of outcome. This is not surprising given their role in disease predisposition rather than prognosis. Prior series, including Pontin and Barnes,12 emphasized that once infection progresses to emphysematous disease, outcomes are more dependent on systemic severity than on underlying risk factors. Our results confirm that aggressive early management can neutralize the effect of these baseline risks. Limitations The main limitation of this study is the modest sample size, which precluded multivariate analysis. Prognostic associations were therefore assessed using univariate methods, which may overestimate certain effects. Nevertheless, the consistent alignment of our findings with large series and meta-analyses strengthens their validity. Summary In summary, our study demonstrates that: 1. The clinical profile and risk factors of EPN in this Indian cohort align with global reports. 2. Minimally invasive drainage achieves high salvage rates even in higher CT grades. 3. Systemic factors—sensorium, dialysis need, hypoalbuminemia, and thrombocytopenia—are stronger prognostic markers than baseline risks or radiological grade. 4. Modern conservative strategies have reduced mortality and nephrectomy rates dramatically, transforming EPN into a condition with favorable outcomes in most patients.

Diabetes mellitus emerged as the predominant risk factor for emphysematous pyelonephritis (EPN); however, prognosis was more closely determined by systemic clinical and biochemical parameters than by baseline comorbidities. Altered sensorium, dialysis requirement, hypoalbuminemia, and thrombocytopenia were the strongest predictors of poor outcome. CT staging continued to provide independent prognostic value and guided therapeutic decision-making, particularly in stratifying the need for intervention. Minimally invasive therapy (DJ stenting with or without PCN/PCD) was highly effective, even in selected patients with higher-grade disease, allowing for high rates of renal salvage. In contrast, extensive surgical interventions such as open drainage and nephrectomy were largely associated with poor outcomes, reflecting advanced disease burden and failure of prior conservative measures rather than ineffectiveness of the surgery itself. Overall, renal salvage with preserved function was achieved in nearly three-fourths of patients, with low nephrectomy (14.3%) and mortality (4.8%) rates. These findings confirm the effectiveness of a stepwise, minimally invasive management approach in the modern era of EPN care. Future Directions • Larger, multicenter prospective studies are needed to validate prognostic models that integrate clinical, biochemical, and radiological parameters. • A robust risk stratification scoring system could aid in timely escalation from conservative to invasive management. • Exploration of novel biomarkers, particularly inflammatory and sepsis-related markers, may improve early risk prediction beyond traditional factors. • Long-term studies should focus on renal functional recovery, dialysis dependence, and health-related quality of life in survivors of EPN.

1. Muñoz-Lumbreras EG, Hernández-Cabezas A, et al. Emphysematous pyelonephritis: A review of its pathophysiology and management. Rev Mex Urol. 2019;79(1):1-13. 2. Ngo XT, Tran AT, Pham VH, et al. Prevalence and risk factors of mortality in emphysematous pyelonephritis patients: A meta-analysis. World J Surg. 2022;46(10):2377-88. doi:10.1007/s00268-022-06634-5. 3. Ubee SS, McGlynn L, Fordham M. Emphysematous pyelonephritis. BJU Int. 2011;107(9):1474-78. doi:10.1111/j.1464-410X.2010.09660.x. 4. Wan YL, Lo SK, Bullard MJ, Chang PL. Predictors of outcome in emphysematous pyelonephritis. J Urol. 1996;156(5):1567-71. doi:10.1016/S0022-5347(01)65482-3. 5. Storey B, Paterson R, Ali H, et al. Emphysematous pyelonephritis: A twelve-year review in a regional centre. J Urol Surg. 2022;9(3):203-8. doi:10.4274/jus.galenos.2022.2022.0002. 6. Somani BK, Nabi G, Thorpe P, Hussey J, Cook J, N’Dow J. Is percutaneous drainage the new gold standard in the management of emphysematous pyelonephritis? Evidence from a systematic review. J Urol. 2008;179(5):1844-9. doi:10.1016/j.juro.2008.01.019. 7. Huang JJ, Tseng CC. Emphysematous pyelonephritis: Clinicoradiological classification, management, prognosis, and pathogenesis. Arch Intern Med. 8. Mohsin N, Budruddin M, Lala S, Al-Taie S. Emphysematous pyelonephritis: A case report and review of the literature. Saudi J Kidney Dis Transpl. 2002;13(2):194-9. 9. Menon M, Devi G, Sanjay D. Emphysematous pyelonephritis. Urol Clin North Am. 2000;27(4):815-20. doi:10.1016/S0094-0143(05)70122-0. 10. Huang JJ, Tseng CC, et al. Updated insights into emphysematous pyelonephritis: A multicenter study. J Urol Adv. 2022;42(3):197-204. 11. Reddy VV, Rao KP. A clinical study of emphysematous pyelonephritis. IAIM. 2018;5(2):150-9. 12. Pontin AR, Barnes RD, Joffe J, Kahn D. Emphysematous pyelonephritis in diabetic patients. Br J Urol. 1995 Jan;75(1):71-4. PMID: 7539599. 13. Sokhal AK, Kumar M, Purkait B, et al. Emphysematous pyelonephritis: Changing trend of clinical spectrum, pathogenesis, management and outcome. Turk J Urol. 2017;43(2):202-9. doi:10.5152/tud.2017.93054.