Urolithiasis DOI 10.1007/s00240-014-0747-z
ORIGINAL PAPER
Percutaneous nephrolithotomy for infection stones: what is the risk for postoperative sepsis? A retrospective cohort study Ohad Shoshany · David Margel · Camil Finz · Orly Ben‑Yehuda · Pinhas M. Livne · Ronen Holand · David Lifshitz
Received: 24 March 2014 / Accepted: 22 December 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract The purpose of this study was to compare the postoperative course of patients with positive stone culture (SC) to patients with sterile SC, and to analyze the predictors for post percutaneous nephrolithotomy (PCNL) sepsis; and to describe the type and resistance patterns of bacteria responsible for post PCNL sepsis. From a cohort of 206 consecutive patients undergoing PCNL, we identified 45 patients with a positive SC (group A) and compared them to patients with a sterile SC (group B). Association between different groups was assessed using Chi square, two-tailed Student’s t test and Mann–Whitney U test, as appropriate. Twenty-three patients had postoperative sepsis and regression analysis was performed to identify clinical variables associated with sepsis. Demographics, stone load and hospitalization time were similar in both groups. Postoperative sepsis developed in 31.1 % of patients in group A compared to 5.9 % in group B. In multivariate analysis, only positive SC was an independent risk factor for postoperative sepsis (OR 6.894, 95 % CI 2.31–20.59, P = 0.001). All patients responded well to treatment with no septic complications. Enterococci were the prevalent organism (29.4 %) in patients with a positive SC. Quinolone resistance was high in both gram negative and gram positive bacteria. Patients with an infected stone are at high risk to develop postoperative sepsis despite standard preoperative antibiotic preparation. SC is important to direct further treatment as almost half of patients with positive SC have a discordant or sterile urine culture. Determining the
O. Shoshany (*) · D. Margel · C. Finz · O. Ben‑Yehuda · P. M. Livne · R. Holand · D. Lifshitz Urology Department, Rabin Medical Center, Petach Tikva and Sakler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel e-mail: [email protected]
prevalent bacteria and resistance patterns in SC can aid the selection of empiric antibiotic therapy in high-risk patients. Keywords Stone culture · Sepsis · Percutaneous nephrolithotomy
Introduction Percutaneous nephrolithotomy (PCNL) is the treatment of choice for large renal calculi, particularly in patients with infection stones, as even minute residual fragments can form a nidus for further infection. Struvite stones occur in association with urinary infection by urea-splitting bacteria. However, any calcified or non-calcified stone may become colonized by infectious organisms associated with an incidental urinary tract infection [1]. Postoperative sepsis is one of the most common complications after PCNL, with an incidence reported between 9.8 and 37 %, while severe sepsis and septic shock occur in 0.3–4.7 %, with risk of mortality [2, 3, 6, 9–16]. There is little data in the literature regarding the specific risk of postoperative infection in patients defined as having infected stones in comparison to non-infected patients. Clinically, an infected stone is suspected when a renal stone is detected in patients with recurrent urinary tract infections (UTI), particularly with persistent bacteriuria. However, the only absolute proof for an infected stone is bacterial growth in a stone culture (SC). In our center, SCs are performed routinely during PCNL and therefore we could identify a group of patients with bacterial growth in a SC, and define them as having an infected stone. The purpose of the current study was to analyze postPCNL infectious complications in our cohort of patients, to compare the postoperative course of patients with positive SC to patients with sterile SC, and to analyze the predictors
13
Urolithiasis
for post-PCNL sepsis. Furthermore, we describe the different SC bacteria, their prevalence and resistance patterns for commonly used antibiotics.
Materials and methods After obtaining institutional review board approval, we queried our prospective PCNL database from January 2008 to December 2012. From 206 consecutive patients, we identified a group of 45 patients with bacterial growth in SC obtained during PCNL (group A). We compared their characteristics and results with patients having a sterile SC (group B). Furthermore, we identified 23 patients (11.2 %) who had postoperative sepsis and analyzed the association of clinical variables with the risk of developing postoperative sepsis. Preoperative data included record of previous history of UTI, prior urine culture (UC) results and renal stones assessment by a non-contrast computed tomography. A preoperative UC was obtained for all patients, 2–4 weeks prior to surgery. All patients with a negative preoperative UC received prophylactic antibiotics with 1 g cefamezin in accordance with the AUA/EAU guidelines [21, 30]. Patients with a positive UC or a history compatible with an infection stone were treated preoperatively with a full course of antibiotics for a minimum of 7 days, adjusted according to the results of the most recent positive UC, usually with quinolones when applicable. Two surgeons (DL and RH) performed all PCNLs. PCNL was performed according to a standard prone technique, using balloon dilator to 30F and an Amplatz sheath. Stone disintegration was performed with ultrasound lithotripsy and ballistic lithotripsy, as required. Flexible nephroscopy was extensively used in all patients to treat residual stones. A nephrostomy tube was routinely placed at the end of the procedure to allow a secondary procedure when required. A sample of fragmented stones was collected, and the surface contaminates were washed off using the standardized Stamey method [4]. Crushed stones were cultured on thiosulfate citrate bile salt sucrose and MacConkey’s agar. All patients underwent routine imaging postoperatively. A nephrostogram was performed routinely before removal of the nephrostomy tube (usually on postoperative day 2) to assess for extravasation, passage of urine into the bladder, and in cases of residual stones to better define the stone location. Stone-free status, defined as no residual stones of 3 mm or larger, was assessed before discharge by abdominal X-ray in patients with radiopaque stones, or by abdominal non-contrast CT for radiolucent stones. Postoperative data included routine heart rate, respiratory rate, blood pressure and temperature measurements
13
and complete blood count. Patients were monitored closely in the postoperative period to watch for signs of systemic inflammatory response syndrome (SIRS), defined as the development of two of four criteria, namely fever less than 36 °C or more than 38 °C, heart rate more than 100 beats per minute, respiratory rate more than 20 breaths per minute and white cell count more than 12,000/ml or less than 4,000/ml [5]. Before defining sepsis, the patients were treated with hydration and analgesics, and other causes for SIRS such as phlebitis or atelectasis were ruled out by physical examination and chest X-ray. Patients developing sepsis were treated according to a recent positive UC, or empirically with gentamicin and ampicillin. Adjustments to the postoperative antibiotic treatment were made, if indicated, once SC results were available, usually within 3 days after the operation. The association between different groups was assessed using the Chi square, two-tailed Student’s t-test and Mann– Whitney U test, as appropriate. Logistic regression modeling was used to analyze the association between clinical variables and postoperative sepsis. All statistical tests were two-sided and for all comparisons P
ORIGINAL PAPER
Percutaneous nephrolithotomy for infection stones: what is the risk for postoperative sepsis? A retrospective cohort study Ohad Shoshany · David Margel · Camil Finz · Orly Ben‑Yehuda · Pinhas M. Livne · Ronen Holand · David Lifshitz
Received: 24 March 2014 / Accepted: 22 December 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract The purpose of this study was to compare the postoperative course of patients with positive stone culture (SC) to patients with sterile SC, and to analyze the predictors for post percutaneous nephrolithotomy (PCNL) sepsis; and to describe the type and resistance patterns of bacteria responsible for post PCNL sepsis. From a cohort of 206 consecutive patients undergoing PCNL, we identified 45 patients with a positive SC (group A) and compared them to patients with a sterile SC (group B). Association between different groups was assessed using Chi square, two-tailed Student’s t test and Mann–Whitney U test, as appropriate. Twenty-three patients had postoperative sepsis and regression analysis was performed to identify clinical variables associated with sepsis. Demographics, stone load and hospitalization time were similar in both groups. Postoperative sepsis developed in 31.1 % of patients in group A compared to 5.9 % in group B. In multivariate analysis, only positive SC was an independent risk factor for postoperative sepsis (OR 6.894, 95 % CI 2.31–20.59, P = 0.001). All patients responded well to treatment with no septic complications. Enterococci were the prevalent organism (29.4 %) in patients with a positive SC. Quinolone resistance was high in both gram negative and gram positive bacteria. Patients with an infected stone are at high risk to develop postoperative sepsis despite standard preoperative antibiotic preparation. SC is important to direct further treatment as almost half of patients with positive SC have a discordant or sterile urine culture. Determining the
O. Shoshany (*) · D. Margel · C. Finz · O. Ben‑Yehuda · P. M. Livne · R. Holand · D. Lifshitz Urology Department, Rabin Medical Center, Petach Tikva and Sakler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel e-mail: [email protected]
prevalent bacteria and resistance patterns in SC can aid the selection of empiric antibiotic therapy in high-risk patients. Keywords Stone culture · Sepsis · Percutaneous nephrolithotomy
Introduction Percutaneous nephrolithotomy (PCNL) is the treatment of choice for large renal calculi, particularly in patients with infection stones, as even minute residual fragments can form a nidus for further infection. Struvite stones occur in association with urinary infection by urea-splitting bacteria. However, any calcified or non-calcified stone may become colonized by infectious organisms associated with an incidental urinary tract infection [1]. Postoperative sepsis is one of the most common complications after PCNL, with an incidence reported between 9.8 and 37 %, while severe sepsis and septic shock occur in 0.3–4.7 %, with risk of mortality [2, 3, 6, 9–16]. There is little data in the literature regarding the specific risk of postoperative infection in patients defined as having infected stones in comparison to non-infected patients. Clinically, an infected stone is suspected when a renal stone is detected in patients with recurrent urinary tract infections (UTI), particularly with persistent bacteriuria. However, the only absolute proof for an infected stone is bacterial growth in a stone culture (SC). In our center, SCs are performed routinely during PCNL and therefore we could identify a group of patients with bacterial growth in a SC, and define them as having an infected stone. The purpose of the current study was to analyze postPCNL infectious complications in our cohort of patients, to compare the postoperative course of patients with positive SC to patients with sterile SC, and to analyze the predictors
13
Urolithiasis
for post-PCNL sepsis. Furthermore, we describe the different SC bacteria, their prevalence and resistance patterns for commonly used antibiotics.
Materials and methods After obtaining institutional review board approval, we queried our prospective PCNL database from January 2008 to December 2012. From 206 consecutive patients, we identified a group of 45 patients with bacterial growth in SC obtained during PCNL (group A). We compared their characteristics and results with patients having a sterile SC (group B). Furthermore, we identified 23 patients (11.2 %) who had postoperative sepsis and analyzed the association of clinical variables with the risk of developing postoperative sepsis. Preoperative data included record of previous history of UTI, prior urine culture (UC) results and renal stones assessment by a non-contrast computed tomography. A preoperative UC was obtained for all patients, 2–4 weeks prior to surgery. All patients with a negative preoperative UC received prophylactic antibiotics with 1 g cefamezin in accordance with the AUA/EAU guidelines [21, 30]. Patients with a positive UC or a history compatible with an infection stone were treated preoperatively with a full course of antibiotics for a minimum of 7 days, adjusted according to the results of the most recent positive UC, usually with quinolones when applicable. Two surgeons (DL and RH) performed all PCNLs. PCNL was performed according to a standard prone technique, using balloon dilator to 30F and an Amplatz sheath. Stone disintegration was performed with ultrasound lithotripsy and ballistic lithotripsy, as required. Flexible nephroscopy was extensively used in all patients to treat residual stones. A nephrostomy tube was routinely placed at the end of the procedure to allow a secondary procedure when required. A sample of fragmented stones was collected, and the surface contaminates were washed off using the standardized Stamey method [4]. Crushed stones were cultured on thiosulfate citrate bile salt sucrose and MacConkey’s agar. All patients underwent routine imaging postoperatively. A nephrostogram was performed routinely before removal of the nephrostomy tube (usually on postoperative day 2) to assess for extravasation, passage of urine into the bladder, and in cases of residual stones to better define the stone location. Stone-free status, defined as no residual stones of 3 mm or larger, was assessed before discharge by abdominal X-ray in patients with radiopaque stones, or by abdominal non-contrast CT for radiolucent stones. Postoperative data included routine heart rate, respiratory rate, blood pressure and temperature measurements
13
and complete blood count. Patients were monitored closely in the postoperative period to watch for signs of systemic inflammatory response syndrome (SIRS), defined as the development of two of four criteria, namely fever less than 36 °C or more than 38 °C, heart rate more than 100 beats per minute, respiratory rate more than 20 breaths per minute and white cell count more than 12,000/ml or less than 4,000/ml [5]. Before defining sepsis, the patients were treated with hydration and analgesics, and other causes for SIRS such as phlebitis or atelectasis were ruled out by physical examination and chest X-ray. Patients developing sepsis were treated according to a recent positive UC, or empirically with gentamicin and ampicillin. Adjustments to the postoperative antibiotic treatment were made, if indicated, once SC results were available, usually within 3 days after the operation. The association between different groups was assessed using the Chi square, two-tailed Student’s t-test and Mann– Whitney U test, as appropriate. Logistic regression modeling was used to analyze the association between clinical variables and postoperative sepsis. All statistical tests were two-sided and for all comparisons P
