Analgesic Use and Complications Following Upper Pole Access for Percutaneous Nephrolithotomy

Abstract

Introduction and Objectives:

Upper pole access (UPA) to the kidney optimizes working angles during percutaneous nephrolithotomy (PCNL). However, many surgeons are hesitant to employ UPA due to concerns for increased postoperative pain and complications. In this study, we compare perioperative pain and outcomes for PCNL performed through UPA with middle and lower pole access (LPA).

Methods:

A retrospective review of PCNLs performed at a single academic institution between 2002 and 2012 was performed. Patient demographics, stone burden, operative characteristics, length of hospital stay, and perioperative complications were recorded. Narcotics administered on postoperative days (POD) 1 and 2 were converted to intravenous morphine equivalents. The Mann–Whitney U, Fisher's exact test, and logistic regression analyses were used for statistical comparison.

Results:

Of 350 PCNLs performed on 269 patients, 125 patients underwent primary PCNL through single tract UPA, while primary LPA was utilized in 138 patients. Patients with multiple access tracts (n=6) were excluded. Baseline characteristics between groups were similar except for age; patients with UPA were older (52.6 vs. 46.2 years; p=0.003). Stone burden, operative time, and initial stone-free rate were comparable between groups. Additionally, analgesic requirements were similar on POD 1 (p=0.54) and 2 (p=0.62). There was no difference in the overall complications (p=0.23) or thoracic complications (p=0.43). Although both groups had equivalent stone-free rates after initial PCNL (p=0.78), those with residual stones had a superior stone-free rate after second-look PCNL through UPA (73.3% vs. 44.3%, p=0.035) and a superior overall stone-free rate (94.4% vs. 86.2%, p=0.037).

Conclusions:

Patients undergoing PCNL through UPA had a higher ultimate stone-free rate compared with LPA, while postoperative analgesic requirements and overall complication rates were similar between access sites. A trend toward higher thoracic complication rates was observed in the UPA group, although this did not reach statistical significance. Surgeons should not hesitate to utilize UPA when indicated for patients with large or complex stones.

Introduction

Effective renal stone clearance for percutaneous nephrolithotomy (PCNL) is contingent upon access of the appropriate calyx. Upper pole access (UPA) provides optimal alignment with most of the intrarenal collecting system and the upper ureter and, therefore, has been utilized in the treatment of complex stone disease. It has been effectively used for staghorn calculi, large or complex calyceal calculi, calculi located at the ureteropelvic junction or in the upper ureter, calculi in anomalous kidneys, and calculi in morbidly obese patients.^1

–4

Due to the normal anatomic relationships between the kidney, lung, and visceral organs, access into the upper pole calyx may be associated with risks of pleural effusion, pneumothorax, and laceration of the lung, liver, and spleen.^5
–7 At our institution, UPA is employed while avoiding access above the 11th rib to minimize thoracic complications. In addition, ultrasound is selectively utilized in patients with complex anatomic abnormalities to establish an access tract that avoids injury to the intra-abdominal organs, pleura, and lung.

Another concern associated with the use of UPA is increased analgesic requirement secondary to pain from irritation of the intercostal nerve.⁸ Patients who experience increased pain may be hesitant to routinely perform full inspiration, further increasing the risk of splinting. This in turn leads to a higher incidence of atelectasis, pleural effusion, and pneumonia.⁹ These adverse events may prolong patients' hospital stay and increase perioperative morbidity. The purpose of this study is to compare the outcomes of PCNL in patients with UPA as compared with those who received middle or lower pole access (LPA).

Patients and Methods

A retrospective review of all PCNLs performed at a single academic institution between 2002 and 2012 was performed. Patients were grouped into two groups: those with UPA and those with middle or LPA. Patient demographics, operative characteristics, and perioperative outcomes were collected for all patients. Demographics collected included age, gender, and body mass index (BMI). Stone burden was assessed by a preoperative computed tomography (CT) scan or abdominal X-ray. To calculate stone burden, the largest two dimensions of each stone were multiplied to yield a cross-sectional area. In patients with multiple stones, the individual cross-sectional areas were summed to determine the total stone burden. Operative characteristics included operative time, fluoroscopy time, and estimated blood loss. The fluoroscopy time reported in this series includes only fluoroscopy utilized intraoperatively for tract dilation, stone removal, and drain placement. During the time period studied, fluoroscopy required for placement of the initial percutaneous access in the interventional radiology suite was not consistently available. Perioperative outcomes included length of postoperative hospital stay, transfusion rate, stone-free rate after the primary PCNL, stone-free rate after secondary PCNL, and ultimate stone-free rate. Perioperative complications were also recorded as Clavien-Dindo scores as described by de la Rosette et al.¹⁰ Postoperative pneumothorax, symptomatic pulmonary edema, pneumonia, and pleural effusion requiring supplemental oxygen or drain placement were recorded as thoracic complications. Finally, patients' inpatient postoperative narcotic use was recorded for both groups. All narcotics were converted to intravenous morphine equivalents using a standard equianalgesic dose chart¹¹ and reported for postoperative days 1 and 2.

Statistical comparison was accomplished using univariate analysis and subsequent logistic regression (IBM SPSS statistics version 20), with ultimate stone-free rate, analgesic use, and overall complication rates as the primary outcome variables. Continuous variables were compared using the Mann-Whitney U test, and categorical variables were compared with the Fisher's Exact test. Results were considered statistically significant at p<0.05.

Technique

Percutaneous renal access was performed with the patient in the prone position generally under IV sedation. Access was established by an experienced interventional radiologist after consultation with the attending urologic surgeon. UPA was routinely attempted in patients with staghorn calculi extending to the upper pole, large-volume ureteral stones, patients with horseshoe kidneys, and patients with isolated upper pole stone burden. If UPA could not be obtained without puncture above the 11th rib, alternate access was used since placing a LPA was felt to be safer for the patient than placing access above the 11th rib. In patients with kidneys located in a cephalad position or patients with significant anatomic abnormalities, ultrasound was used at operator discretion to avoid injury to the intra-abdominal organs, pleura, and lung.

PCNL was performed with the patient in a prone position under general anesthesia. The nephrostomy tract was dilated to 30F using a NephroMax balloon (Boston Scientific, Natick, MA), and stone removal was achieved using ultrasonic and laser lithotripsy combined with manual extraction. Endoscopic mapping was performed with a flexible nephroscope (Karl Storz, Tuttlingen, Germany) to ensure complete stone removal before termination of the procedure. At the conclusion of the procedure, a soft 22F Council tip nephrostomy tube (Bard Medical, Covington, GA) was placed and secured to the patient's skin. In addition, a 5F re-entry catheter was placed into the bladder.

Postoperative oral narcotics were routinely used for primary pain control, with intravenous narcotics utilized for breakthrough pain. Nonsteroidal anti-inflammatory medications were not used in the acute postoperative period due to the risk of bleeding. On the first postoperative day (POD 1), patients underwent abdominal and pelvic CT to screen for residual stone fragments. Residual calculi 4 mm in diameter or greater were considered significant. Significant residual calculi were treated with second-look PCNL, shock wave lithotripsy (SWL), or ureterorenoscopy at the discretion of the treating physician. Following final PCNL, patients underwent clamping trial and subsequent nephrostomy tube removal if the clamp trial was successful.

Results

During the study period, 350 PCNLs were performed on 269 patients. Patients undergoing PCNL through multiple access tracts (6 patients undergoing 9 PCNLs) were excluded, leaving 263 patients for analysis. There were 125 patients who underwent primary PCNL with UPA, 34 patients with midpole calyceal access, and 104 patients with lower pole calyceal access. The remaining 78 PCNLs consisted of second-look or subsequent PCNLs performed for residual stone burden. Patient characteristics and operative outcomes based on univariate analysis are described in Table 1. When the primary outcomes of the stone-free rate, complication rate, and postoperative narcotic use were compared, there were no differences between patients with middle and lower pole accesses, so these were grouped together for the purposes of analysis. Adjunctive ultrasound guidance was utilized while establishing percutaneous renal access in 53.2% of UPA and 42.6% of LPA (p=0.18). Baseline characteristics between groups were similar except for age; patients with UPA were older, with mean age 52.6 versus 46.2 years (p=0.003). Stone burden was large, but similar between groups (UPA 736.2 mm², LPA 752.4 mm², p=0.45), and there was no difference in the proportion of staghorn calculi (partial and full) treated by each approach (UPA 54.5%, LPA 56.9%, p=0.71). Only 1.5% of our cohort underwent PCNL for isolated ureteral stones. The operative time (p=0.45), estimated blood loss (p=0.70), and postoperative hospital stay (p=0.98) were also similar. Additionally, mean morphine-equivalent analgesic requirements were similar on postoperative day 1 (18.2 mg for UPA, 18.4 mg for LPA, p=0.54) and 2 (16.4 mg for UPA, 17.1 mg for LPA, p=0.62).

Table 1.

Patient Characteristics, Operative Data, and Outcomes of Upper Pole Access and Lower Pole Access Groups

Characteristic	Upper pole access (n=125)	Middle and lower pole access (n=138)	p-Value
Mean age (years)	52.6	46.2	0.003
Number male/female	70/55	64/74	0.14
Mean BMI (kg/m²)	30.4	30.6	0.81
Mean stone burden (mm²)	736.2	752.4	0.45
Staghorn stones	54.4%	56.9%	0.71
Laterality (left/right)	71/54	68/70	0.27
Ultrasound-assisted access	53.2%	42.6%	0.18
Mean operative time (minute)	143.2	137.1	0.45
Mean estimated blood loss (mL)	114.1	140.8	0.70
Mean fluoroscopy time (second)	124.8	124.0	0.13
Mean postoperative hospital stay (days)	4.0	4.4	0.98
Transfusion rate	4.0%	3.6%	1.00
First-look stone-free rate	92/125 (73.6%)	99/138 (71.7%)	0.78
Second-look stone-free rate	22/30 (73.3%)	13/30 (44.3%)	0.035
Ultimate stone-free rate	118/125 (94.4%)	119/138 (86.2%)	0.037
Overall complication rate	25.6%	18.8%	0.23
Thoracic complication rate	3.2%	1.4%	0.43
Mean morphine-equivalent analgesic use, postoperative day 1 (mg)	18.2	18.4	0.54
Mean morphine-equivalent analgesic use, postoperative day 2 (mg)	16.4	17.1	0.62

Values in bold are statistically significant.

BMI=body mass index.

After initial PCNL, 73.6% of UPA patients and 71.7% of LPA patients were rendered stone free (p=0.78). However, patients with UPA who required second-look PCNL had a significantly higher stone-free rate for the second procedure as compared with patients with LPA (22/30 vs. 13/30, p=0.035). After taking into account all procedures performed, patients with UPA ultimately achieved a higher stone-free rate (94.4% vs. 86.2%, p=0.037). Using logistic regression, the correlation between percutaneous access location and ultimate stone-free rate remained significant after adjustment for age, BMI, and stone burden (p=0.035).

There was no significant difference in the overall postoperative complication rates (UPA 25.6% vs. LPA 18.8%, p=0.23) on univariate analysis. After performing logistic regression to control for age and BMI, no statistically significant difference in complication rates was observed (OR 1.60 for complications with UPA, 95% CI 0.86–2.98, p=0.141).

The Clavien-Dindo score distribution is displayed in Table 2. Minor complications (Clavien-Dindo 1) included fever, self-limited hematuria, and pain requiring additional opiate control such as patient-controlled analgesia. Moderate complications (Clavien-Dindo 2) included pleural effusion managed conservatively and blood loss necessitating transfusion. Clavien-Dindo 3 complications included thoracic drain placement, secondary cystoscopy, or ureteral stent placement.

Table 2.

Clavien-Dindo Score Distribution

Clavien-Dindo score	Upper pole access (n=125)	Middle and lower pole access (n=138)
1	14	10
2	9	10
3A	6	2
3B	—	—
4A	3	4
4B	—	—
5	—	—

Major complications (Clavien-Dindo 4A) in the UPA group included one patient with uncontrolled hypertension, one patient who suffered a postoperative myocardial infarction, which was medically managed, and one patient with arterial bleeding requiring embolization; this same patient also developed acute pancreatitis, which led to a prolonged 26-day hospital stay. Among the LPA group, two patients required reintubation, and two patients developed significant intraoperative acute blood loss anemia with one developing an arrhythmia requiring cardiopulmonary resuscitation and transfusion. All patients with major complications were treated in the intensive care unit.

Six patients developed thoracic complications after initial PCNL: four in the UPA group and two in the LPA group (3.2% vs. 1.4%, p=0.43). In the UPA group, two patients developed a pleural effusion requiring supplemental oxygen and two sustained a pneumothorax. One patient required only supplemental oxygen for treatment of his pneumothorax, which was detected on the postoperative CT scan. The other patient had no evidence of pneumothorax on the postoperative CT scan, but instead presented to the emergency department with dyspnea on POD 8. The chest X-ray showed a left pneumothorax, which was managed by placement of an 8.5F chest drain. In the LPA group, there was one patient with a pleural effusion requiring oxygen and one patient who developed pneumonia and was reintubated. Patients who experienced thoracic complications showed a trend toward increased postoperative length of stay compared with those without thoracic complications (9.0 days vs. 4.1 days, p=0.10). However, one patient with a pleural effusion also had a prolonged postoperative hospital stay of 24 days due to unrelated gastrointestinal symptoms. No patient sustained intra-abdominal organ injury, and there were no perioperative deaths.

Discussion

In a normally oriented kidney, the upper pole lies posteromedial to the lower pole of the kidney. Hence, the upper pole approach provides a straight tract along the long axis of the kidney and allows access to most of the collecting system, while providing easier manipulation with rigid instruments.^9,12 This is thought to be advantageous in the treatment of large or complex renal and upper ureteral calculi. In our population, an equivalent stone-free rate was achieved after primary PCNL regardless of access location. However, patients who were not rendered stone-free after the primary procedure had a significantly better stone-free rate after a secondary PCNL, if it was performed through UPA as opposed to LPA (73.3% vs. 44.3%, p=0.035). This in turn led to the superior ultimate stone-free rate seen in the UPA group (94.4% vs 86.2%, p=0.037). We hypothesize that the improved visualization and angles of access derived from UPA may be an important factor in achieving complete stone clearance in patients with a larger stone burden or complex anatomy.

However, upper pole PCNL can also be associated with increased complications. In a recent multicenter analysis of outcomes after PCNL, the complication rates from UPA and LPA have been reported as 23.5% and 16.1%, respectively.¹³ Our data showed a similar complication rate of 25.6% and 18.8%, respectively; however, there are several important differences in our patient population. Patients in our series had a higher stone burden than is seen in the Clinical Research Office of the Endourological Society (CROES) database, with the UPA stone burden 736.2 mm² versus 476.7 mm² and LPA 752.4 mm² versus 442.1 mm². This likely reflects the propensity of our center to treat smaller stones with retrograde ureterorenoscopy or SWL and utilize PCNL for larger or more complex stones. In addition, our operative times were longer than in the CROES database (UPA 143.2 minutes vs. 92.4 minutes and LPA 137.1 minutes vs. 75.1 minutes), again likely due to a higher operative complexity in our patients. We noted a 3.2% incidence of thoracic complications in UPA compared with 1.4% in the LPA group, which was not statistically different (p=0.43). This was similar to the reported hydrothorax rates in the CROES database (5.8% for UPA and 1.5% for LPA). Although there was an increased rate of hydrothorax in the UPA group in the CROES database, their subgroup analysis showed that there was no difference in hydrothorax rates for subcostal versus supra-12th rib access.¹³ The patients who received supra-11th rib access did have a significantly higher rate of hydrothorax; however, at our institution, we do not place supra-11th rib access, potentially explaining why our cohort showed no difference in thoracic complications by renal access location.

To access the upper pole of the kidney, supracostal puncture is typically done between the 11th and 12th rib, ideally with the patient in full expiration. Due to an increased risk of damage to the lung and subsequent thoracic complications,¹² we avoid puncture above the 11th rib. Although the supracostal approach also has potential to lacerate the spleen, liver, and bowel,^5
–7,14 no injury to visceral organs occurred in our series, possibly because of our selective use of ultrasound in patients at risk for these injuries. In addition, to establish UPA, we place the nephrostomy in a medial position to minimize damage to the liver, spleen, intercostal vessels, and nerves.⁸

In many patients, supracostal access will traverse the pleura and leave the patient at risk for pleural effusion.¹² Pleural effusions can be caused by inadequate tamponade of the nephrostomy tract, inadequate drainage of the kidney after the procedure, and failure to seal the tract with a working sheath during stone removal.¹⁵ Interestingly, no patient in our series required thoracic drain placement for pleural effusion after the initial PCNL, and only one required a drain for pneumothorax. At our institution, the decision to intervene for pleural effusion or pneumothorax is based on clinical parameters with confirmatory radiographic imaging. Patients with mild symptoms may be treated with conservative measures such as supplemental oxygen, incentive spirometry, and diuresis. However, patients with more severe symptoms may require pleural drainage. This was accomplished with small-bore 8.5F pleural drains, which have a similar efficacy as large-bore chest tubes for the treatment of pleural effusion and pneumothorax, with lower morbidity.¹⁶ If thoracic drainage is required, small-bore pleural drains should be considered.

One of the primary outcomes of this study was the quantification of postoperative pain associated with UPA versus LPA by using objective morphine equivalents of opiates administered on postoperative days 1 and 2. We found that postoperative pain management was not significantly different between groups, although there was significant variability in opiate requirements (range 0–113 mg IV morphine equivalents for the entire cohort on POD 1 and 0–130 mg IV morphine equivalents on POD 2). This is in agreement with the subset analysis of Kirac et al., who showed no difference in postoperative analgesic requirements in patients with supra- versus subcostal percutaneous access. However, in their study, this was not a primary outcome, and analgesic use included both opiate and nonopiate medication.¹⁷ To our knowledge, this is the first study that directly compares postoperative opiate use in patients with differing renal accesses for PCNL.

Traditionally, postoperative pain is believed to be greater for supracostal access because the diaphragm is punctured and each breath taken results in irritation of the diaphragm,¹⁸ which may be exacerbated by a nephrostomy tube left in place postoperatively. Indeed, one study showed a 32% rate of chest pain after supracostal puncture versus 5% in the subcostal group, although they did not analyze whether patients required pain medication or determine the amount of narcotics required.⁵ Puncture of the parietal pleura may be another potential source of pain with supracostal upper pole renal access. One study used CT scans of healthy volunteers in inspiration and expiration to assess the risk of pleural violation with a theoretical supra-12th rib UPA; they estimated that there would be a 29% chance of traversing the pleura on the right side and 14% chance on the left.¹² However, multiple studies have shown a clinical rate of thoracic complications of 5–7%^6,13 for supracostal or UPA. The exact incidence of upper pole supracostal puncture of the pleura and its contribution to postoperative pain has not been well defined.

However, our results show that patients with UPA and LPA had equivalent opiate requirements, suggesting that additional variables may play a greater role in postoperative pain than renal access location. We propose several mechanisms to explain this finding. First, we attempt to place supracostal access tracts in the middle of the intercostal space. Cadaveric dissections have shown that the anatomical position of the 11th intercostal nerve puts it at risk for injury during supra-12th rib percutaneous access,⁸ and nephrostomy placement at the inferior border of the rib should be avoided to minimize the risk of nerve injury. However, if the nephrostomy is placed at the superior border of the 12th rib, the forces applied to the access sheath intraoperatively as well as the sliding of the nephrostomy tube postoperatively may cause periosteal irritation, leading to greater postoperative pain. In addition, greater forces required to insert the sheath over the rib and subsequent deformation of the sheath may increase parenchymal damage to the kidney. By placing the nephrostomy in the middle of the intercostal space, we avoid intercostal nerve injury and minimize the local trauma and irritation to the 12th rib.

Second, we use ultrasound guidance in patients with difficult anatomy, including patients with severe scoliosis, contractures, renal anomalies, or in patients whose preoperative CT suggests complex anatomic relationships. This allows real-time monitoring of the access needle during nephrostomy placement. Continuous visualization of the access tract may allow the physician to avoid injury to the pleura and intra-abdominal structures, thereby reducing postoperative pain from pleural puncture as well as inadvertent laceration of the bowel and other organs.

Finally, we use a meticulous surgical technique, including liberal use of flexible nephroscopy, to reduce the trauma associated with torquing of the rigid nephroscope. By minimizing the torque applied to the kidney intraoperatively, we aim to minimize tearing of the renal capsule and parenchyma. Because UPA provides more direct alignment with the intrarenal collecting system and ureter, less manipulation may be required to treat large or complex calculi with UPA than with LPA. We hypothesize that the surgical technique during PCNL affects postoperative pain and subsequent narcotic requirements.

A variety of interventions have been studied in an attempt to reduce pain and analgesic requirements after PCNL. Grimsby et al.¹⁹ evaluated the effect of continuous intravenous ketorolac versus placebo in patients undergoing renal surgery, including laparoscopic donor nephrectomy and PCNL; in their study, ketorolac did not reduce the pain scores postoperatively. However, other interventions, such as administration of intravenous paracetamol,²⁰ infiltration of the nephrostomy tract with bupivicaine,²¹ and intercostal nerve block,²² have all demonstrated a reduction in perioperative analgesic use after PCNL.

There have been a number of studies published on the optimal nephrostomy tube size after PCNL. Pietrow et al. randomized patients to a 10F pigtail catheter or 22F Council nephrostomy after PCNL. Patients who received the smaller nephrostomy catheter had lower pain scores at 6 hours after surgery, but no difference in pain scores or narcotic use was seen on POD 1 and 2.²³ In addition, a recent analysis of a multinational PCNL database showed that large-bore (>18F) nephrostomy tubes were associated with a lower complication rate and less blood loss than small-bore tubes.²⁴ In light of the decreased complications and no difference in postoperative pain associated with large-bore nephrostomy drainage, we feel that the use of the 22F Council catheter is justified.

Tubeless PCNL has been studied extensively and advocated by some, as a method to reduce postoperative pain.²⁵ However, a tubeless technique does not allow a second-look procedure if residual stone fragments are present, and at our institution, patients who are stone free after PCNL generally have their nephrostomy removed in 24–48 hours after surgery. In addition, by leaving a large bore, nephrostomy patients, who require second-look PCNL, can undergo flexible nephroscopy without additional dilation of the percutaneous access tract.

Our study does have several limitations, including the retrospective nature, moderate cohort size, and single institution study design. Recent articles have analyzed several of the same outcomes utilizing an international database; however, these results may not be generalizable to the United States patient population, as the authors mention in their discussion.¹³ Additionally, in our study, narcotic use was used as an objective measure for postoperative pain. This was only analyzed for the first 2 days after PCNL, as a standardized postoperative care algorithm was used for these days. Analgesic requirement associated with initial placement of the nephrostomy tube was not recorded. The type and duration of oral analgesic use after discharge were also not available, allowing us to comment only about the initial postoperative pain control. Further studies utilizing a prospective design may help elucidate the optimal strategy for maximizing both safety and efficacy during PCNL.

Conclusions

Overall complication rates and postoperative analgesic requirements were not higher in patients undergoing PCNL with upper pole renal access. However, a nonsignificant trend toward higher thoracic complications in the UPA group was noted. UPA patients ultimately achieved a superior stone-free rate, potentially due to improved angles of access to the entire collecting system. Surgeons should not hesitate to utilize UPA for patients with large or anatomically complex stones when clinically indicated.

Footnotes

Acknowledgment

The authors would like to thank Udochukwu Oyoyo for his assistance with statistical analysis.

Disclosure Statement

Dr. D. Duane Baldwin serves as a lecturer for Cook Medical and as a consultant for Terumo. No competing financial interests exist for any of the other authors.

Abbreviations Used

References

Wolf

Jr. , Clayman

. Percutaneous nephrostolithotomy: What is its role in 1997?. Urol Clin North Am, 1997; 24:43–58.

Stening

, Bourne

. Supracostal percutaneous nephrolithotomy for upper pole caliceal calculi. J Endourol, 1998; 12:359–362.

Golijanin

D KR

, Verstandig

, Sasson

, Landau

, Meretyk

. The supracostal percutaneous nephrostomy for treatment of staghorn and complex kidney stones. J Endourol, 1998; 12:403–405.

Lojanapiwat

, Prasopsuk

. Upper-pole access for percutaneous nephrolithotomy: Comparison of supracostal and infracostal approaches. J Endourol, 2006; 20:491–494.

Radecka

E BM

, Holmgren

, Magnusson

. Complications associated with percutaneous nephrolithotripsy: Supra- versus subcostal access. Acta Radiol, 2003; 44:447–451.

Munver

R DF

, Newman

, Preminger

. Critical analysis of supracostal access for percutaneous renal surgery. J Urol, 2001; 166:1242–1246.

Muzrakchi

AA SW

, Omar

, Younes

. Is the 10th and 11th intercostal space a safe approach for percutaneous nephrostomy and nephrolithotomy?. Cardiovasc Intervent Radiol, 2004; 27:503–506.

McAllister

MLK

, Torrey

, Chenoweth

, Barker

, Baldwin

. Intercostal vessels and nerves are at risk for injury during supracostal percutaneous nephrostolithotomy. J Urol, 2011; 185:329–334.

Lawrence

, Cornell

, Smetana

; American College of Physicians. Strategies to reduce postoperative pulmonary complications after noncardiothoracic surgery: Systematic review for the American College of Physicians. Ann Intern Med, 2006; 144:596–608.

10.

de la Rosette

, Opondo

, Daels

, et al. Categorisation of complications and validation of the Clavien score for percutaneous nephrolithotomy. Eur Urol, 2012; 62:246–255.

11.

Pasero

, McCaffery

. Pain Assessment and Pharmacologic Management. St Louis, Missouri: Mosby Elsevier, 2011, pp. 444–445.

12.

Hopper

, Yakes

. The posterior intercostal approach for percutaneous renal procedures: Risk of puncturing the lung, spleen, and liver as determined by CT. Am J Roentgenol, 1990; 154:115–117.

13.

Tefekli

, Esen

, Olbert

, et al. Isolated upper pole access in percutaneous nephrolithotomy: A large-scale analysis from the CROES percutaneous nephrolithotomy global study. J Urol, 2013; 189:568–573.

14.

Gupta

, Kumar

, Kapoor

, Srivastava

, Mandhani

. Prospective evaluation of safety and efficacy of the supracostal approach for percutaneous nephrolithotomy. BJU Int, 2002; 90:809–813.

15.

Picus

, Weyman

, Clayman

, McClennan

. Intercostal-space nephrostomy for percutaneous stone removal. Am J Roentgenol, 1986; 147:393–397.

16.

Fysh

, Smith

, Lee

. Optimal chest drain size: The rise of the small-bore pleural catheter. Semin Respir Crit Care Med, 2010; 31:760–768.

17.

Kirac

, Tepeler

, Bozkurt

, et al. The efficacy of bupivacaine infiltration on the nephrostomy tract in tubeless and standard percutaneous nephrolithotomy: A prospective, randomized, multicenter study. Urology, 2013; 82:526–531.

18.

Namasivayam

, Gayathri

Indu

. Perioperative respiratory compromise in patients undergoing PCNL-A case series. J Clin Diagn Res, 2013; 7:2310–2312.

19.

Grimsby

, Conley

, Trentman

, et al. A double-blind randomized controlled trial of continuous intravenous Ketorolac vs placebo for adjuvant pain control after renal surgery. Mayo Clinic Proc, 2012; 87:1089–1097.

20.

Maghsoudi

, Tabatabai

, Radfar

, et al. Opioid sparing effect of intravenous paracetamol after percutaneous nephrolithotomy (PCNL): A double-blind randomized controlled trial. J Endourol, 2014; 28:23–27.

21.

Shah

, Shah

, Sodha

, Khandkar

, Gokhale

. A randomized control trial evaluating efficacy of nephrostomy tract infiltration with bupivacaine after tubeless percutaneous nephrolithotomy. J Endourol, 2012; 26:478–483.

22.

Honey

, Ghiculete

, Ray

, Pace

. A randomized, double-blinded, placebo-controlled trial of intercostal nerve block after percutaneous nephrolithotomy. J Endourol, 2013; 27:415–419.

23.

Pietrow

, Auge

, Lallas

, et al. Pain after percutaneous nephrolithotomy: Impact of nephrostomy tube size. J Endourol, 2003; 17:411–414.

24.

Cormio

, Preminger

, Saussine

, et al. Nephrostomy in percutaneous nephrolithotomy (PCNL): Does nephrostomy tube size matter? Results from The Global PCNL Study from The Clinical Research Office Endourology Society. World J Urol, 2013; 31:1563–1568.

25.

Desai

, Kukreja

, Desai

, et al. A prospective randomized comparison of type of nephrostomy drainage following percutaneous nephrostolithotomy: Large bore versus small bore versus tubeless. J Urol, 2004; 172:565–567.