Emphysematous Pyelonephritis (EPN) is a life-threatening infection, characterized by gas formation within or around the kidney [1-5]. The prevalence of the disease is very low in the urology practices in western countries, however, the incidence is more in the Indian subcontinent due to more cases of uncontrolled diabetes mellitus [2]. EPN is more commonly seen in diabetics with poor glycemic control and very occasionally in pelvicalyceal obstruction due to stone diseases [6,7]. It is most commonly caused by gas-forming organisms, most commonly Escherichia coli, Klebsiella and very rarely, case reports have been published with Clostridium, Candida, Aspergillus, Cryptococcus and Amoeba [1,6-9]. The clinical presentation of EPN is similar to uncomplicated pyelonephritis, however, it has an aggressive course with high morbidity and mortality rate. The treatment of EPN is controversial, traditionally early nephrectomy was considered the treatment of choice in emphysematous pyelonephritis and the mortality rate was around (60-90%) [1,5,7]. Currently the management of EPN is changing to a more conservative approach like early diagnosis, early control of blood sugar, appropriate antibiotics, Percutaneous Nephrostomy drainage (PCN) and good supportive care [4-13].The mortality has reduced to 21% with the above measures and early nephrectomy should be avoided [10,13,14, 15].

The aim and objectives of our study were to elucidate the clinical features, evaluate the prognostic factors of EPN and outcome among the various radiological classes of EPN.

It was a single-center retrospective study, conducted in the department of urology RIMS Imphal Manipur, from January 2016 to August 2020. The study was conducted after taking appropriate consent from the ethics board committee of the institute. The consent was taken from all the study participants. The medical records of the patients were reviewed from the department of medical record section of the urology ward with a diagnosed case of emphysematous pyelonephritis.

After collection of the file records, the following data were analyzed:

• Demography (Age, sex and underlying diagnosis)

• The site of involvement whether unilateral or bilateral 

• Clinical features at initial presentation (hemodynamic status, shock, degree of consciousness, etc.)

• Biochemical parameters (leukocyte counts, Platelet count, HbA1c (glycated hemoglobin) and serum creatinine, etc. 

• Class of disease as defined by CT scan (Huang and Tseng) [1]

• Organism cultured from urine and PCN fluid

• Outcome among various radiological classes of EP and mortality among different groups

After recording the above findings, the study participants were divided into, cured and expired groups to elucidate the risk factors. All the patients were treated in the department of urology and the regular help of a nephrologist, physician, endocrinologist and anesthesiologist were taken when required. The standard treatment protocols for fluid resuscitation, insulin therapy as well as antibiotic therapy were followed as per the department protocols. Empirical antibiotic treatment was started with 3rd generation cephalosporins and/or carbapenem group and the further course of antibiotics was changed depending on the microbiological culture reports of blood/urine sample.

Definitions

Following are the various definitions used in our study.

Emphysematous Pyelonephritis

Presence of culture-positive UTI and presence of gas in the renal parenchyma/collecting system as evident on CT scan.

Class of Disease

Classification of the disease was based on Huang and Tseng's classification [1]. 

Sepsis [1, 15]

The presence of two or more of the following conditions as a result of infection:

• Temperature >38°C or <36°C

• Heart rate >90 beats/ minute

• Respiratory rate >20 breaths/minute or PaCO₂ <32 mmHg and white blood cell count (>12,000/mm³ or <4,000/mm³) or >10% immature (band) forms

Thrombocytopenia, Shock and DIC [14, 15]

Thrombocytopenia was defined as a platelet count of <1,20,000. Shock was defined as blood pressure of <100/60 mmHg at admission. DIC was defined as elevated prothrombin time, elevated activated partial thromboplastin time (aPTT), reduced platelet count and low fibrinogen level. In cases where thrombocytopenia and DIC co-existed, thrombocytopenia was considered as a part of DIC.

Statistical Analysis

Descriptive analysis was done for various clinical parameters. All data were expressed as a percentage, mean±standard deviation. To test the predictors of poor prognosis, the Fisher exact test (two-tailed) was used for categorical variables. A p-value of <0.05 was considered sign.

There were 20 patients with a mean age of 46.25±9.76. Out of 20, 12(60%) patients were female and 8(40%) patients were male. 18 cases were diabetic and 2 cases had impacted upper ureteric stone. The left side was involved in 11(55%) cases, the right side in 6(30%) and 3(15%) cases were bilaterally involved. One patient with left-sided EPN had A solitary kidney Table 1.

Table 1: Baseline Characteristics of Patients 

#One patient with left-sided was solitary kidney

Flank pain 19(95%), fever 17(85%) and dysuria 9(45%) were the most common symptoms. Altered sensorium was present in 3(15%) cases. Sepsis was present in 16(80%) cases, shock in 5(25%) cases, DIC (disseminated intravascular coagulation) in 3(15%) cases and WBC count was raised in 18(90%) cases with a mean of 22,045/µL (±7608.40). HbA1c was raised in 17(85%) cases with a mean of 7.69% (±1.30). The most common organism isolated was E. coli 14(70%) and the other organism cultured were Klebsiella (15%), one candida, one acinetobacter, one staphylococcus aureus Table 2.

Table 2: Clinical and laboratory examination findings of patients

*disseminated intravascular coagulation, #one acinetobacter, one candida, and one staphylococcus aureus

Based on Huang and Tseng classification [1]. Out of 20 cases, 4(20%) cases were class 1, all were managed conservatively with no mortality. About 8(40%) cases were class 2 managed conservatively and PCN was kept in 4 patients and all got treated well (Figure 1). 

Figure 1: Class 2 Left emphysematous pyelonephritis (Computed tomography scan showing gas (arrowhead) in left kidney Parenchyma)

There were 4(20%) cases in class 3, out of which 2 cases underwent PCN with medical management and one patient underwent nephrectomy who died after 2 days of operation (Figure 2). 

Figure 2: Class 3A Right emphysematous pyelonephritis. (Computed tomography scan showing gas in right kidney parenchyma (arrowhead) extending into the perinephric area)

About 4(20%) cases were class 4, out of which 2 patients died and 2 patients survived/cured Table 3 and Figure 3.

Figure 3: (Class 4) Bilateral emphysematous pyelonephritis (Computed tomography scan in axial and coronal sections showing left side grade 1 gas involving only PCS (red arrowhead) and right-sided extensive EPN with the extension of gas into the perinephric space, pararenal area and extending downward to iliopsoas muscle and right thigh (black arrowhead)

Table 3: Management and outcome in Patients with Emphysematous Pyelonephritis based on Huang and Tseng Classification [1]

Emphysematous Pyelonephritis (EPN), Percutaneous Nephrostomy (PCN), *Four cases were of class 4(3 cases bilateral EPN and one left solitary kidney) 

We found high class (3 and 4) based on Huang and Tseng classification, shock on admission, DIC and serum creatinine ≥5mg/dL were the poor prognostic factors Table 4.

Table 4: Findings of Prognostic Factors in Study Groups (Survival Vs. Mortality Group)

*Statistically Significant Difference (p-value<0.05); p-value by Fisher’s exact test

EPN is an uncommon life-threatening condition characterized by the production of gases within the renal parenchyma and perirenal space [1-5] Our analysis shows that the disease was mostly had a female preponderance with a ratio of 2:3 and occurred mostly in middle age groups (40-50) with a mean age of (46.25±9.76) [1,3,6,7,9]. The left kidney (55%) was frequently involved than the right kidney (30%).Huang and Tseng have postulated that four factors are involved in the pathogenesis of EPN, including gas-forming bacteria, high blood glucose level, impaired tissue perfusion and a defective immune response (such as DM) [1].

In our study 18 patients out of 20(90%) were diabetic and HbA1c was raised in 17(85%) patients who had poor control of their blood sugar before getting EPN and multiple studies had shown similar findings [5,6,9]. Two patients (10%) had EPN due to upper ureteric obstruction due to long-standing impacted calculus, managed by emergency PCN drainage and both the patients recovered well and elective ureter lithotomy was done after 3 months. Some of the case studies also have shown urinary tract obstruction (5.8-60%) as the other predisposing condition [5,7].

We found E. coli in 14(70%) patients, Klebsiella in 3(15%) patients, 1 acinetobacter, 1 Candida and 1 Staphylococcus aureus case. In a systematic review of published data on EPN, they found E. coli was the culprit in 65.6% of cases, Klebsiella in 19.5% of cases and mixed organism in 10% [16,11].

All the patients in this case series were first managed in casuality where adequate resuscitation was done. Early control of blood sugar was done by giving short-acting regular insulin and an opinion from the physician was taken for the proper management of blood sugar. The type of antibiotics was given as per the institutional protocol and a further course of antibiotics was added as per the urine and PCN culture report. All the patients in grade 1&2 were managed conservatively with no mortality.

All the patients with classes 1 and 2 were managed conservatively and cured. One patient in class 3 underwent immediate nephrectomy and the patient died on the 2nd postoperative day. The patient went into diabetic ketoacidosis, sepsis, electrolyte imbalances, acute respiratory acidosis and the patient died due to cardiopulmonary arrest. Kapoor et al. [7] also found early nephrectomy is associated with higher mortality rates than is initial conservative management. Two patients died in class 4 groups and both patients had uncontrolled blood sugar with multiple poor prognostics factors.

Flank pain 19(95%) and fever 17(85%) were the most common symptoms. In a study of 48 patients of EPN, Huang et al reported fever in (79%) and flank pain (71%) as the most common symptoms [1] and the same was also observed by Tang et al. [17] in their series of 21 patients.

Sepsis was involved in 80% of cases of the survived group, whereas in the mortality group sepsis was present in all of the cases. Surprisingly we did not find sepsis as a poor prognostic factor in the outcome of cases with or without sepsis (p = 1.000) and this indifference maybe because of less number of cases in our study.

In our study thrombocytopenia was present in 3(15%) cases, out of which 2(66.6%) cases were present in the expired group (p-value 0.045). thrombocytopenia at admission has been reported to be an independent prognostic factor previously as well [1,14]. However, since thrombocytopenia was present in cases that had DIC, we have considered it to be part of DIC and therefore have not discussed it as an independent risk factor. Patients in DIC had a higher mortality (p = 0.045). Hypotension was seen in five (25%) patients of our study and we found an independent predictor of poor outcome (p = 0.009). We found serum creatinine of >5mg/dL as a predictor of the poor outcome (p-value 0.031) [1,14]. Our study shows significant differences in the clinical features among the four classes, there was a tendency towards a higher mortality rate from class 1 to 4 EPN (p-value- 0.049). In patients with (class 3 and 4), we found the presence of ≥2 risk factors as a poor predictor of the disease outcome (p-value-0.004) [1,2]. 

The advantages of PCD include drainage of pus, relief of gas pressure to local circulation and providing a high success rate in extensive EPN. In all of the cases, the percutaneous drainage was done using a 12-14 F pigtail catheter. 

The overall mortality in our series was 15%, much less than what has been previously reported. There has been a decrease in mortality in the last two decades due to early detection, adequate resuscitation, early control of blood sugar and most importantly the widespread adoption of percutaneous drainage [2,5,6,8].

The main limitation of our study was being a retrospective analysis and the number of cases was few to analyze the other risk factors.

E. coli is the commonest organism responsible for the disease. Early control of blood sugar with vigorous hydration and appropriate intravenous extended-spectrum antibiotic should be the initial treatment of choice. Early percutaneous nephrostomy should be tried even in advanced stages. Shock, DIC and Serum creatinine >5mg/dL on admission are independent predictors of poor outcome.

1. Huang, J.J. and C.C. Tseng. “Emphysematous pyelonephritis: Clinicoradiological classification, management, prognosis and pathogenesis.” Archives of Internal Medicine, vol. 27, 2000, pp. 797–805. https://doi.org/10.1001/archinte.160.6.797.

2. Bhat, R.A. et al.“Emphysematous pyelonephritis: outcome with conservative management.” Indian Journal of Nephrology, vol. 23, no. 6, pp. 444–447.

3. Wan, Y.L. et al. “Acute gas-producing bacterial renal infection: correlation between imaging findings and clinical outcome.” Radiology, vol. 198, 1996, pp. 433–438.

4. Flores, G. et al. “Acute bilateral emphysematous pyelonephritis successfully managed by medical therapy alone: A case report and review of the literature.” BMJ Nephrology, vol. 3, 2002, p. 4. https://doi.org/10.1186/1471-2369-3-4.

5. Ambar, K. et al. “Retrospective analysis of clinical profile, prognostic factors and outcomes of 19 patients of emphysematous pyelonephritis.” International Urology and Nephrology, vol. 41, 2009, pp. 959–966. https://doi.org/10.1007/s11255-009-9552-.

6. Uruc, F. et al. “Emphysematous pyelonephritis: our experience in managing these cases.” Canadian Urological Association Journal, vol. 9, nos. 7–8, 2015.

7. Kapoor, R. et al. “Predictive factors for mortality and need for nephrectomy in patients with emphysematous pyelonephritis.” BJU International, vol. 105, 2010, pp. 986–989.

8. Falagas, M.E. et al. “Risk factors for mortality in patients with emphysematous pyelonephritis: A meta-analysis.” Journal of Urology, vol. 178, 2007, pp. 880–885. https://doi.org/10.1016/j.juro.2007.05.017.

9. Stapleton, A. “Urinary tract infections in patients with diabetes.” American Journal of Medicine, vol. 113, 2002, pp. 80–84. https://doi.org/10.1016/S0002-9343(02)01062-8.

10. Kuchay, M.S. et al.“Medical therapy alone can be sufficient for bilateral emphysematous pyelonephritis: Report of a new case and review of previous experiences.” International Urology and Nephrology, vol. 46, 2014, pp. 223–227.

11. Liao, H.W. et al.“Emphysematous pyelonephritis caused by Bacteroides fragilis.” Nephrology Dialysis Transplantation, vol. 20, 2005, pp. 2575–2577. https://doi.org/10.1093/ndt/gfi146.

12. Cardinael, A.S. et al. “Emphysematous pyelonephritis: Successful treatment with percutaneous drainage.” American Journal of Roentgenology, vol. 164, 1995, pp. 1554–1555.

13. Tahir, H. et al. “Successful medical treatment of acute bilateral emphysematous pyelonephritis.” American Journal of Kidney Diseases, vol. 36, 2000, pp. 1267–1270.

14. Lu, Y.C. et al. “Predictors of failure of conservative treatment among patients with emphysematous pyelonephritis.” BMC Infectious Diseases, vol. 14, 2014, p. 418.

15. Angus, D.C. and T. van der Poll. “Severe Sepsis and septic shock.” New England Journal of Medicine, vol. 369, no. 9, 2013, pp. 840–851.

16. Somani, B.K. et al.“Is Percutaneous drainage the new gold standard in the management of emphysematous pyelonephritis? Evidence from a systematic review.” Journal of Urology, vol. 179, 2008, pp. 1844–1849. https://doi.org/10.1016/j.juro.2008.01.019.

17. Tang, H.J. et al.“Clinical characteristics of emphysematous pyelonephritis.” Journal of Microbiology, Immunology and Infection, vol. 34, 2001, pp. 125–130.