Supracostal percutaneous nephrolithotomy: A prospective comparative study.

INTRODUCTION: A widely prevalent fear of thoracic complications with the supracostal approach has led to its underutilization in percutaneous nephrolithotomy (PCNL). We frequently use the supracostal approach and compared the efficacy and thoracic complications of infracostal, supra 12(th), and supra 11(th) punctures.
MATERIALS AND METHODS: This was a prospective study of patients who underwent PCNL between January 2005 and December 2012. The patients were divided into three groups based on the access: infracostal, supra 12(th) (between the 11(th) and 12(th) ribs) and supra 11(th) (between the 10(th) and 11(th) ribs). Clearance rates, fall in hemoglobin levels, transfusion rates, perioperative analgesic requirements, hospital stay and thoracic complications were compared.
RESULTS: Seven hundred patients were included for analysis. There were 179 (25.5%) patients in the supra 11(th) group, 187 (26.7%) patients in the supra 12(th) group and 334 (47.8%) patients in the infracostal group. The overall clearance rate was 78% with no difference in the three groups. The postoperative analgesic requirements were significantly higher in the supracostal groups and showed a graded increase from infracostal to supra 12(th) to supra 11(th). During the study period, only 2 patients required angioembolization (0.3%) and none required open exploration. The number of patients requiring intercostal chest drain insertion was extremely low, at 1.6% and 2.2% in the supra 12(th) and supra 11(th) groups, respectively.
CONCLUSIONS: Our results confirm the feasibility of the supracostal approach including punctures above the 11(th) rib, albeit at the cost of an increase in thoracic complications. Staying in the line of the calyx has helped us to minimize the most dreaded complication of bleeding requiring angioembolization.

INTRODUCTION

An ideal percutaneous nephrolithotomy (PCNL) puncture has been described as one that provides the shortest and straightest access to all calculi, avoids major vessels, bowel and lung, lies along the axis of the calyx and causes minimal parenchymal damage. Sampaio, in his elegant description of the calyceal anatomy, has emphasized the need for a puncture to go through the fornix rather than the infundibulum to avoid hemorrhage.[1]

Current clinical practice continues to reflect a reluctance to use the supracostal approach. Supracostal access in general and supra 11th access in particular, continues to be underutilized due to an unfounded fear of thoracic complications. In the large, multicenter, transcontinental Clinical Research Office of the Endourological Society (CROES) PCNL Study report, access above the 11th rib was used only in 1.5–2.2% patients and 68–83% punctures were infracostal.[2] The assumpion seems to be that the higher the access; the higher will be the complication rate. We, however, believe that the ease of access to the pelvicalyceal system (PCS) offered by the supracostal access counteracts any potential thoracic complications. We report our outcomes and complications with various sites of access for PCNL.

MATERIALS AND METHODS

This prospective study, conducted between January 2005 and December 2012, included all patients who underwent PCNL and was approved by our Institutional Review Board. For this report, patients were divided into three groups based on the access being infracostal, supra 12th (between the 11th and 12th) ribs, and supra 11th ribs (between the 10th and 11th). Functional study (intravenous urography) was done in all patients after the confirmation of the stones by computerized tomography (CT) scan or ultrasound. Exclusion criteria included concomitant supra and infracostal access, kidneys with abnormal anatomy, severe kyphoscoliosis, and previous thoracic surgeries on the ipsilateral side and patients with other surgeries done simultaneously which could have affected the study. All patients received general anesthesia and a single dose of 4 mg epidural morphine.

Percutaneous nephrolithotomy technique

In the prone position, access to the desired calyx was planned at the discretion of the individual surgeon. The essential principle followed was that the access tract should be in line with the axis of the calyx, and the tract should split the papilla. If a line along the axis of the calyx was projecting supracostally then, the supracostal approach was preferred.

With the fluoroscopy C-arm in the 0° position, the edge of the calyx was marked on the skin. A line along the axis of the calyx was marked on the skin. A point approximately 5–7 cm from the edge of the calyx was taken as the site of the puncture. The needle remained in this axis while the depth of the puncture was decided. The needle was introduced up to the edge of the calyx on 0°. The needle depth was then adjusted to ensure that in both 0 and 30°, the needle was equidistant from and pointing at the center of the calyx. The calyx was punctured in 0° with a slight to and fro movement to confirm the entry through the desired point.

All attempts were made to introduce a hydrophilic guide wire down the ureter and coil it in the bladder. A balloon dilator was used to dilate the tract, and a 30F Amplatz sheath was introduced. A characteristic popping sound can often be heard, when the Amplatz sheath passes through the pleura. A 28 Fr nephrostomy tube was left indwelling for 48 h in supracostal access but was removed in 24 h after infracostal access. We believe that this allows the tract to mature and pleurorenal connections to close. Postoperative pain assessment was done using the visual analog scale in adults and Wong Baker Face Scale in children. Breakthrough pain relief was by intravenous tramadol.

Clearance rates, fall in hemoglobin levels, transfusion rates, perioperative analgesic requirements, hospital stay and thoracic complications were compared in these groups. Stone clearance was documented by postoperative radiographs and ultrasound done by a single ultrasonologist. Residual fragments of any size were considered as failures. Hemoglobin levels were done on the 1st postoperative day and compared with the preoperative levels.

Descriptive and inferential statistical analysis was carried out. Results on continuous variables were presented as mean ± standard deviation (min-max) and the results on categorical variables were presented as numbers (%). Significance was assessed at 5% level of significance. Analysis of variance was used to find the significance of study parameters between three and more groups of patients. The Chi-square test was used to find the significance of study parameters on a categorical scale between two and more groups. A multivariate logistic regression analysis was performed to find the prediction of morbidity using clinical variables.

RESULTS

A total of 777 patients underwent PCNL during the study period. After exclusion, 700 patients were included for analysis. There were 179 (25.5%) patients in the supra 11th group, 187 (26.7%) patients in the supra 12th group and 334 (47.8%) patients in the infracostal group. The three groups were well matched in terms of age, renal function, stone size, the number of tracts and presence of absence of positive preoperative urinary cultures [Table 1].

Table 1

Comparison of baseline clinical variables in three groups of patients studied

Outcome variables are presented in Table 2. The overall clearance rate was 78% with no difference in the three groups. Postoperative analgesic requirements were significantly higher in the supracostal groups and showed a graded increase from infracostal to supra 12th to supra 11th. The hospital stay showed the statistical difference but with the mean ranging from 4.18 to 4.87 days this does not seem to have any practical relevance. Blood transfusion was needed in 4.1% of patients with no difference in the hemoglobin fall or transfusion rates among the groups. During the study period, only 2 patients required angioembolization (0.3%) and none required open exploration.

Table 2

Comparison of outcome variables in three groups of patients studied

Table 3 lists the thoracic complications encountered in the study. Overall, complication rates did not show any statistical difference. Thoracic complications showed a graded increase as we progressed from infracostal (1.8%) to supra 11th (15.9%). However, the number of patients requiring intercostal chest drain (ICD) insertion was extremely low, at 1.6% and 2.2% in the supra 12th and supra 11th groups, respectively. Of the 19 patients showing hydrothorax, only 7 required ICD insertion.

Table 3

Comparison of complications in three groups of patients

DISCUSSION

Our study reflects the results of a teaching hospital where surgeons at varying levels along the learning curve perform PCNL. Our overall clearance rates are comparable with results reported by Tefleki in the CROES PCNL study where the stone-free rates in the upper pole access group and lower pole access groups were 77.1% and 81.6% respectively.[2]

This is a large single-center study, and to our knowledge, includes the largest number of supra 11th access reported in the literature. The CROES PCNL study group, reflecting practices across countries, reported the use of access above the 11th rib in only 1.5–2.2% of cases.[3] Seitz meta-analysis of 115 studies on the incidence and prevention of complications categorically states that punctures above the 11th rib should be avoided.[4] Although cadaveric and radiological studies predict a high likelihood of organ injuries and pleural transgression, clinical experience has been to the contrary.[56] In contrast to the CROES data, 25.5% of our access tracts were supra 11th, 26.7% were supra 12th and less than half were infracostal.

The ideal puncture has been described as one that provides the shortest and straightest access to all calculi, avoids major vessels, bowel, and lung, lies along the axis of the calyx and causes minimal parenchymal damage. If we need to make all punctures in the axis of the calyx, it stands to reason that the calyces projecting supracostally should have a supracostal access. With the patient supine and in complete expiration, 58–70% of middle calyces and 20–22% of lower calyces were above the rib. Urologists, of course, are more concerned about the site of puncture rather than the calyx. If one draws a straight line from the edge of the calyx to the skin along the axis of the calyx, a large majority in the upper and interpole calyces, will intersect the skin above the 12th rib. These tracks will naturally be supracostal.

Kekre et al. reported a 9.8% incidence of pleural collections requiring the intercostal chest tube drainage.[7] An early CT determined the risk of adjacent organ injury seemed to have added to the fear of the supracostal approach. Hopper and Yakes reported that the chances of transgressing the lungs in complete expiration were 29% on the left and 14% on the right in a supra 12th approach and almost universal in a supra 11th approach.[5]

Contrary to the above opinion, there have been a number of reports documenting the safety of the supracostal approach. Even three decades ago, there have been authors who have advocated the supracostal puncture.[89] Maheshwari et al. reported that in 150 patients with supra 12th approach there was not even one thoracic violation.[10] In a letter to the editor in response to the article by Kekre et al. they outline the precautions that they take to minimize complications in the supracostal approach. They advocate for making the puncture in the lateral half of the intercostal space; keeping the patient in deep expiration during puncture; always keeping the needle below the 10th rib, dilating the tract only to the minimum size necessary and obtaining a postprocedure chest film.[10] Others have concurred by reporting pleural collection rates in the range of 0–4%.[31112131415] The safety of this approach has also been documented in pediatric[16] and tubeless PCNL. The additional purported advantages of the tubeless approach are reduced pain and a shorter hospital stay.[1718]

In the Mannheim experience of 315 patients, renal hemorrhage requiring angioembolization was reported in 0.3%.[19] Others have reported angioembolization rates of 0.6–1.4%.[2021] In comparison, we had to refer only 2 patients (0.3%) for angioembolization during the study period. Our low rates of hemorrhage can be explained on our insistence that all punctures must remain along the line of the calyx and that every effort must be made by the surgeon to split the papilla and enter through the center of the calyx.

Blood transfusion rates in the three groups in our study ranged from 2.2% to 5.3% with no significant difference (P = 0.307). Michel et al. and Aron et al. reported blood transfusion rates upto 17.5%.[2223] Bleeding from the intercostal arteries has been blamed as a possible etiology for the increased bleeding seen in supracostal punctures seen in earlier series. Our experience suggests that there is no increase in bleeding complications in supracostal access.

We did note a graded increase in thoracic complications on moving from the infracostal to the supra 11th approach. The incidence of hydrothorax increased from 0.3% (infracostal) to 2.7% (supra 12th) to 7.3% (supra 11th). Atelectasis/pneumonitis was seen in 1.5%, 3.7%, and 8.6% respectively. ICD was required in 0%, 1.6%, and 2.2% respectively. The small percentage of patients with hydrothorax brings into question our practice of doing a chest X-ray for all patients with a supracostal puncture after ICD removal. Others have recommended chest X-ray only in symptomatic patients.[7]

More importantly, the need for ICD insertion was limited to 7 patients (1%) throughout the study. This is in spite of our liberal use of the supracostal approach. This is in stark contrast to Munver's report where there was a 23.1% intrathoracic complication rate in supra 11th punctures and 1.4% in supra 12th punctures.[24] Understandably, there were no patients requiring intercostal drainage in the infracostal group.

Maheshwari et al. have recommended precautions to prevent intrathoracic complications.[10] One of their recommendations is to make the puncture in deep expiration. Our practice is to allow the normal respiration to continue. We aim to avoid the lung injury and not pleural violation. In normal respiration, the lung margin remains 5 cm above the parietal pleura and access through the pleurodiaphragmatic recess would avoid the lungs.

We, in fact, are often able to demonstrate a characteristic popping sound when the Amplatz sheath is pushed across the pleura over the balloon dilator. Our efforts are directed at preventing thoracic complications, and we follow some universal precautions assuming all our supracostal punctures to be transpleural. These additional precautions, which have helped to keep our intercostal drainage rates to a bare minimum, include:

Ensuring that the Amplatz is securely within the PCS throughout the procedure.

bA quick 3 step dilatation using 2 Teflon dilators followed by balloon dilatation allows less time for air or fluid to get indrawn into the pleural space during dilatation.

Guard against the collar of the nephroscope occluding the outlet of the Amplatz sheath. This happens when a short nephroscope is used to access a distant calculus, e.g. in the upper ureter. Impeding the outflow converts the PCS into a high-pressure system, and fluid then extravasates around the Amplatz sheath into the pleural cavity along the path of least resistance.

Minimize the use of irrigant fluid by a one-handed technique of closing the inflow when flow is unnecessary, e.g. while introducing graspers.

The occasional use of high-pressure inflow to improve visualization is strictly contraindicated in supracostal punctures.

The incidence of nonpleural complications also increases as the level of the track ascends. The presence of the nephrostomy tube in the small intercostal space and the resultant splinting of the respiratory effort may contribute to increased atelectasis and pneumonitis in the supracostal group. Better pain management and vigorous physiotherapy could help to prevent these complications. A tubeless approach to PCNL is also reported to be feasible, but we do not follow this approach.[25]

CONCLUSIONS

There is a graded increase in thoracic complications as one ascends from the infracostal to the supra 11th approach. However, our data suggests that the complete avoidance of the supracostal approach is unnecessary. Following a few simple precautions, intrathoracic complications can be kept to a bare minimum.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.