Peri-Operative Predictors of Postoperative Bleeding and Sepsis after Percutaneous Nephrolithotomy

Abstract

Introduction: 

Percutaneous nephrolithotomy (PCNL) is the most common minimal access method used for the management of large renal and pelvic stones. The two most common complications after PCNL are bleeding and sepsis. In this context, we did this study to find out perioperative predictors of postoperative bleeding and sepsis after PCNL.

Materials and Methods: 

We conducted this study on 110 patients who came for PCNL at our institute from March 1, 2018 to August 31, 2019. Data were collected using a pro forma, and the same analyzed by appropriate statistical methods using the SPSS software.

Results: 

The perioperative factors correlated to postoperative bleeding were stone size, density of stone, Guy's stone score, size of Amplatz sheath used, number of working tracts created, and duration of surgery. On regression analysis factors which were predictive of postoperative bleeding were size of Amplatz sheath and number of working tracts created. Peri-operative factors correlated with postoperative sepsis were body mass index of the patient, preoperative total count, density of stone, preoperative pyuria, preoperative positive urine culture, superior calyceal puncture, postoperative total leukocyte count, and postoperative fever within 24 h. The factors which were predictive of postoperative sepsis were preoperative pyuria, preoperative urine culture, superior calyceal puncture, and postoperative fever on regression analysis.

Conclusion: 

This study shows perioperative predictors of postoperative bleeding and sepsis after PCNL. Out of 43.63% total complications, 20% were major and 23.63% minor complications.

Introduction

Urinary tract is one among the most common organ system prone to have stones. Its surgical management which was previously done by open method is now being done by minimal access methods. Percutaneous nephrolithotomy (PCNL) is currently the most common minimal access method used for the management of large calyceal and renal pelvic stones. The safety and efficacy of PCNL have increased since the eighties when it was first introduced due to the refinement of technology, improved surgical instrumentation, and increasing experience.[¹] PCNL can be performed through access into the upper, middle, or lower calyx of the kidney by either fluoroscopic or ultrasound guidance. Although there has been rapid evolution in sizes of nephroscope resulting in the development of mini PCNL, ultramini PCNL, and microperc and in using various energy sources like pneumatic lithotripters, ultrasonic devices and lasers to fragment stones, significant complications are seen in up to one quarter of PCNL patients who undergoes this surgery.[²³]

Studying the factors which affect the complications of PCNL is important to reduce the rate of complications and to improve the quality of service.[⁴] The two most common complications after PCNL are bleeding requiring transfusion and sepsis. It is in this scenario we are doing this research at our institute to assess the perioperative predictors of postoperative bleeding and sepsis after PCNL.

Materials and Methods

This was a prospective study conducted in a tertiary care hospital in Kerala, India. The objective of the study was to find out perioperative predictors of postoperative bleeding and sepsis after PCNL in patients admitted for this procedure at our institute from March 1, 2018 to August 31, 2019. The inclusion criteria were all those patients admitted for PCNL who were willing to participate in the study and whose Karnofsky performance status was 80 and above. The exclusion criteria were those admitted patients who were not willing to participate in the study and patients with Karnofsky performance status below 80.

According to a previous study conducted by Shakhawan H. A., Said et al. from department of Urology, Sulaymaniyah Surgical Teaching Hospital, Iraq; among 200 patients, 47.5% had different types of complications. Seventeen (8.5%) patients had bleeding which needed blood transfusions and 32 (16%) had bleeding which responded to bed rest and i.v crystalloids. Intraoperative perforations occurred in 32 (16%) and infective complications in 30 (15%) patients.[⁵] Using this information, sample size has been estimated to 106 using the formula (Zα)² PQ/d², where Zα is 1.96 for 5% significance, P is 47.5 and Q is (100-P) which is 52.5, d is precision which is 20% of P. Taking into consideration, drop out cases in between study also, final sample size is 110. The study was started after getting institutional research and ethics committee approval. A detailed consent form for the willingness to participate in the study was obtained from patients.

Information about the patients who are participating in the study will be collected using a pro forma. Patients who fulfill inclusion criteria will be personally interviewed and data will be collected. These patients will undergo blood and urine investigations with ultrasonography of Kidney Ureter Bladder (KUB) and computerized tomography (CT) KUB stone protocol. The pro forma will contain details pertaining to age, sex, IP | OP number, relevant history, and body mass index. It contains preoperative variables such as hemoglobin, total white blood cell count, presence of pyuria, urine bacterial culture status, degree of hydronephrosis, stone size, stone density in Hounsfield units identified by CT scan, and stone complexity Guy's Stone Score (GSS) grade.[⁶] GSS comprises four grades: Grade I is solitary stone in mid/lower pole or solitary stone in the pelvis with simple anatomy; Grade II solitary stone in upper pole or multiple stones in a patient with simple anatomy or a solitary stone in a patient with abnormal anatomy; Grade III multiple stones in a patient with abnormal anatomy or stones in a calyceal diverticulum or partial staghorn calculus; Grade IV is staghorn calculus or any stone in a patient with spina bifida or spinal injury. Intraoperative variables in data collection are size of Amplatz sheath, intraoperative position of stone, site of puncture, number of tracts, and duration of surgery.

This procedure PCNL was done under general anesthesia. Prophylactic antibiotics were given according to the institutional protocol. After placing the patient in the lithotomy position, ureteric catheterization was done in retrograde manner with a 6 F ureteric catheter (Aster Medispro, Bangalore, India) using a rigid cystoscope (30° Karl Storz with 22 F outer sheath, Tuttlingen, Germany). Later, the patient was turned to prone position, and all pressure points were adequately padded. The target calyx was accessed under C-arm fluoroscopy (Shimadzu, Kyoto, Japan) guidance using 16 G initial puncture needle (Aster Medispro, Bangalore, India). The floppy tipped guidewire was then passed into the collecting system through the needle. A working tract was established using a serial metallic dilator system under fluoroscopy control and 28–32 F Amplatz sheath (Rusch Teleflex, Morrisville, USA) was placed over the dilated tract. Through Amplatz sheath, a nephroscope (20 degree R. Wolf with 24 F outer sheath, Illinois, USA) was then placed directly into the kidney. Fragmentation of stones was done using a pneumatic lithotripter (Status medical equipments, Satara, India). Forceps and irrigating fluid were used to remove stone fragments. At the end of the procedure, the ureteric catheter would be retained and the same removed next day if there were no complications. A nephrostomy tube (16 F) was placed in all patients and clamped for 8 h. A Per urethral 16 F Foley's catheter was also placed.

Those patients who underwent either blood transfusion or any adjuvant procedures such as angioembolization or surgery will be considered to have postoperative bleeding. Patients who have systemic inflammatory response syndrome (SIRS) with infection or suspected infection (Sepsis 2 criteria) after surgery will be considered in postoperative sepsis category.[⁷] SIRS means two or more of the following such as leukocytosis or leukopenia, hyperthermia or hypothermia, tachycardia, and tachyapnea. The data thus collected will be statistically analyzed to identify whether these perioperative predictive factors are significantly related to postoperative bleeding and sepsis.

Qualitative variables were summarized using percentages and quantitative variables using mean with standard deviation. Then, statistical analysis was done to identify peri-operative factors with correlation to postoperative bleeding and sepsis. Qualitative variables were tested by using Chi-square test and quantitative variables by Pearson's correlation analysis. Those peri operative factors either by Chi-square test or by Pearson's correlation analysis with P < 0.05 were considered significant. Later regression analysis was done to find out significant perioperative predictive factors for postoperative bleeding and sepsis. Perioperative factors with regression co-efficient whose P value is <0.05 at 95% confidence interval (CI) were considered predictive of complications. All statistical analysis was performed using the SPSS software (SPSS Inc. Released 2009. PASW statistics for windows, version 18.0 SPSS Inc., Chicago, Illinois, USA).

Results

In our study, 110 patients underwent PCNL in 18 months. Sex distribution of the study population is shown in Figure 1. The distribution of various preoperative variables is shown in Figures 2–4.

Figure 1: Sex distribution of study population (male = 81 patients and female = 29 patients)

Figure 2: Distribution of preoperative pyuria among patients. (without pyuria = 64 patients, with pyuria = 46 patients)

Figure 3: Distribution of preoperative urine culture result among patients with pyuria

Figure 4: Degree of hydronephrosis by ultrasonography

The total number of patients with complications (both minor and major) was 48 (43.63%). Out of these 26 (23.63%) patients had minor complications such as mild postoperative blood stain in urine (11 patients, 10%) and chest infections (15 patients, 13.64%) which were managed conservatively.

Twenty-two patients (20%) had severe complications such as postoperative bleeding which needed either blood transfusion or adjuvant procedures, sepsis, pulmonary edema, and pleural injury. We had 7 (6.36%) patients with postoperative bleeding which needed further treatment. Out of these seven patients, 4 (3.63%) cases were managed by blood transfusion, 2 (1.81%) cases by doing angioembolization by endovascular coiling and one case (0.90%) was managed by emergency open surgery with suturing of PCNL track in the kidney within 12 h of surgery.

Urinary sepsis was found in 12 patients (10.90%) which were managed by giving antibiotics according to urine culture and sensitivity. Eight patients had E. coli and 4 had Klebsiella bacteria isolated from their urine. In our study population, two (1.81%) patients had postoperative atelectasis which was managed conservatively. One (0.90%) patient developed pulmonary edema due to myocardial infarction within 24 h of surgery and died.

When Pearson's correlation co-efficient analysis was done after analyzing results, perioperative factors correlated to postoperative bleeding [Table 1] were stone size (P = 0.0433), density of stone (P = 0.0131), GSS (P = 0.0119), size of Amplatz sheath used (P = 0.0001), number of working tracts created (P < 0.001), and duration of surgery (P < 0.001). When regression analysis [Table 2] was done, those factors which were predictive of postoperative bleeding were size of Amplatz sheath (95% CI, P = 0.010) and number of working tracts (95% CI, P = 0.002) created only.

Table 1: Correlation analysis of variables for postoperative bleeding

Table 2: Regression analysis showing predictive factors of postoperative bleeding

For postoperative sepsis [Table 3 and Table 4], factors correlated were body mass index of patient (P = 0.0036), preoperative total count (P = 0.0043), density of stone (P = 0.0412), preoperative pyuria (P < 0.001), preoperative positive urine culture (P < 0.001), superior calyceal puncture (P = 0.012), postoperative total leucocyte count at 24 h of surgery (P < 0.0001) and postoperative fever within 24 h (P < 0.0001). When regression analysis was done [Table 5], those factors which were predictive of postoperative sepsis were preoperative pyuria (95% CI, P = 0.032), preoperative urine culture (95% CI, P = 0.030), superior calyceal puncture (95% CI, P = 0.002), and postoperative fever (95% CI, P = 0.001).

Table 3: Correlation analysis of quantitative variables for postoperative sepsis

Table 4: Correlation analysis of qualitative variables for postoperative sepsis

Table 5: Regression analysis showing predictive factors of postoperative sepsis

Discussion

According to review of literature done in this particular topic from PubMed, Embase, and science direct database, total complication rate varies between 28% and 49.8%. In our study, total complication rate (including major and minor) was 43.63%. Rate of major complication in literature was between 7% and 17.8%. Our major complication rate was 20% like postoperative bleeding which needed either blood transfusion or adjuvant procedures, sepsis, pulmonary edema, and pleural injury. We had 23.63% of minor complications. Here in the literature, the rate of minor complications varied between 18.8% and 28.3%.

In a study by Olvera-Posada D et al. which was a series of 2318 patients, blood transfusion rate was 1.4%.[⁸] Blood transfusion rate in our study was 3.63%. After reviewing various other studies, blood transfusion rate varies between 1% and 10%. Among our study population, 1.81% was the rate of angioembolization. According to a study by Arora A M, et al., 0.51% of patients who underwent PCNL at their institute required embolization to control bleeding.[⁹]

According to the literature, predictive factors of postoperative bleeding which needed either blood transfusion or adjuvant procedures were number of tracts dilated, stone type, diabetes, preoperative hemoglobin level, duration of surgery, degree of hydronephrosis, and diabetes. In our study, perioperative factors correlated to postoperative bleeding were stone size (P = 0.0433), density of stone (P = 0.0131), GSS (P = 0.0119), size of Amplatz sheath used (P = 0.0001), number of working tracts created (P < 0.001), and duration of surgery (P < 0.001). Those factors which were predictive of postoperative bleeding were size of Amplatz sheath (95% CI, P = 0.010) and number of working tracts created (95% CI, P = 0.002) only.

Out of 110 patients, 10.90% patients had urinary sepsis, and on literature review, it can vary between 6.2% and 28.9%. According to a study of 241 cases by Liu J, et al., urosepsis occurred in 17% of patients after PCNL. Based on multivariate logistic regression analysis, the independent risk factors associated with postoperative urosepsis included preoperative leukocyte elevation (OR = 3.973, P = 0.005), positive urine nitrite (odds ratio [OR] = 3.697, P = 0.010), and positive urine culture (OR = 3.562, P = 0.002).[¹⁰] In a study by Rivera M, et al. 16% had infectious complications. Eleven patients (5%) with urinary tract infection/pyelonephritis, 21 patients (9%) with SIRS and 2 (0.9%) with sepsis. There were no significant differences between those with and without infectious complication with regard to age, gender, stone size, presence of diabetes, or procedure duration. Those with infectious complication were more likely to have a positive intraoperative stone culture (P = 0.01), struvite stone composition (P < 0.01), staghorn calculi (P < 0.001), and multiple stones (P = 0.02).[¹¹] Another study by Zhu L, et al. preoperative urinary tract infection (OR = 4.38, 95% CI: 1.15–9.53), multiple access (OR = 5.31, 95% CI: 1.23–10.75), diabetes (OR = 4.97, 95% CI: 1.37–9.86), length of operation ≥60 min (OR = 5.67, 95% CI: 2.24–13.42), estimated blood loss in PCNL ≥500 mL (OR = 2.78, 95% CI: 2.32–3.61) were the independent risk factors associated with postoperative infection.[¹²]

In our study factors related to sepsis were body mass index of patient (P = 0.0036), preoperative total count (P = 0.0043), density of stone (P = 0.0412), preoperative pyuria (P < 0.001), preoperative positive urine culture (P < 0.001), superior calyceal puncture (P = 0.012), postoperative total leukocyte count (P < 0.0001) and postoperative fever within 24 h (P < 0.0001). When regression analysis was done, those factors which were predictive of postoperative sepsis were preoperative pyuria (95% CI, P = 0.032), preoperative urine culture (95% CI, P = 0.030), superior calyceal puncture (95% CI, P = 0.002), and postoperative fever (95% CI, P = 0.001).

Conclusion

This study conducted at our institute shows perioperative predictors of postoperative bleeding and sepsis after PCNL. In this study, perioperative factors correlated to postoperative bleeding were stone size, density of stone, GSS, size of Amplatz sheath used, number of working tracts created, and duration of surgery, but those which were predictive of postoperative bleeding were size of Amplatz sheath and number of working tracts created only.

Perioperative risk factors correlated with sepsis were body mass index of patient, preoperative total count, density of stone, preoperative pyuria, preoperative positive urine culture, superior calyceal puncture, postoperative total leukocyte count, and postoperative fever within 24 h but only preoperative pyuria, preoperative urine culture, superior calyceal puncture, and postoperative fever were found to be predictive of postoperative sepsis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Acknowledgments

Our sincere thanks to all faculty members and postgraduates of our department.