Comparison of Percutaneous Nephrolithotomy Under Spinal Versus General Anesthesia: A Randomized Clinical Trial

Abstract

Purpose:

To evaluate the safety and efficacy of spinal anesthesia compared with general anesthesia in patients who underwent percutaneous nephrolithotomy (PCNL).

Patients and Methods:

One hundred patients with American Society of Anesthesiologists (ASA) score <3 were randomly divided into two groups according to the type of anesthesia. Spinal anesthesia was performed using an injection of 0.25 mg/kg bupivacaine 0.5% in the intrathecal space; no opium (fentanyl) agent was used. All procedures were performed with the patient in the prone position. Stone access was made by using fluoroscopic guidance, and the tract was dilated using a single-stage technique. All patients received a solution including 1 mg/kg morphine in every 100 mL physiologic saline through the volumetric pump during the 3-hour post-PCNL period in the recovery room. Afterward, morphine (0.05 mg/kg) was injected only according to the verbal rating scale greater than 3 after discharge from the recovery room until 24 hours after surgery.

Results:

The two groups were matched by mean age, distribution of stone location, and stone burden. Mean operative time, hospital stay, stone-free rate and mean hemoglobin drop were comparable between the two groups. The rate of complications according to the Clavien grading system was nearly similar in both groups. Mean analgesic requirement during 24 hours after PCNL was 6.8 mg in the spinal group and 13.2 mg in the general group (P<0.001).

Conclusion:

It seems that using spinal anesthesia by intrathecal injection of local anesthetic solutions vs general anesthesia has comparable surgical outcomes and reduces the requirement for analgesia after PCNL in the early postoperative period.

Introduction

P ercutaneous nephrolithotomy (PCNL) is performed routinely under general anesthesia and with patients in the prone position.^1
–3 Urologists have modified their techniques in recent years to improve safety and efficacy and decrease morbidity. PCNL in different positions (prone, supine, prone-flexed, and flank), tubeless PCNL, and PCNL with a different type of anesthesia (spinal, general, and combined spinal-epidural) are some modifications have been described in previous studies.^3

–7

Tracheal tube displacement and neurologic events especially at the time of position change are the important concerns during PCNL under general anesthesia.^8,9 Spinal anesthesia has lower adverse effects and is more cost effective than general anesthesia.^10,11 Likewise, awareness of the patient during the change of position may be the main advantage of spinal to general anesthesia. Previously, a limited number of studies evaluated different surgical parameters between combined spinal-epidural anesthesia (CSEA) and general anesthesia.¹²

We assessed surgical outcomes and analgesic requirement after PCNL in two groups of patients who were divided randomly according to the type of anesthesia (spinal vs general). This randomized study was designed to delineate whether PCNL using spinal anesthesia can be considered as an acceptable alternative to PCNL under general anesthesia.

Patients and Methods

The patients were randomly divided into two groups according to anesthesia type (spinal vs general), and written informed consent was obtained from all the patients. Randomization was performed using a random block allocation method. Patients older than 16 years with renal stones larger than 2 cm and negative urine culture results were included in this study. Exclusion criteria were renal anomaly, skeletal deformity, history of PCNL or open stone surgery, American Society of Anesthesiologists (ASA) physical status ≥3, functional single kidney, history of bleeding disorders, and anticoagulant or antithrombotic medication and addiction to opium and alcohol. Body mass index and height of the patients were not as exclusion criteria.

All patients received 1 g cephazoline 30 minutes before anesthesia. Mean arterial pressure (MAP), heart rate, arterial O₂ saturation, and electrocardiogram were monitored from beginning of anesthesia until discharge from the recovery room (6 hours after operation) every 15 minutes, and then these parameters were recorded every 2 hours until 24 hours after surgery. Postoperative vomiting was defined as more than two times vomiting during 24 hours after PCNL.

General anesthesia

Fentanyl at 2 μg/kg and midazolam at 0.03 mg/kg were administered 2 minutes before induction and after losing verbal contact, induction was initiated by injection of 2 mg/kg propofol and 0.5 mg/kg atracurium (relaxant agent), and finally laryngoscopy and tracheal intubation were attempted. Atracurium as a relaxant agent was monitored using the train of four index. Propofol infusion during general anesthesia was adjusted according to the level of cerebral status index (CSI); 20% increase in propofol infusion rate with CSI level greater than 60 and 20% decrease with CSI level lower than 40. Likewise, remifentanyl was administered during anesthesia according to the blood pressure changes (MAP was recorded every 10 minutes). Patients were excluded from this study when optimal blood pressure establishment necessitated injection of other agents. Propofol and remifentanyl infusion were stopped at the end of surgery and 0.05 mg/kg morphine injected to relieve postoperative pain.

Spinal anesthesia

Patients were placed initially in the lateral position and then 0.25 mg/kg bupivacaine 0.5% (up to 40 mg) was injected in the intrathecal space (L3–L4). The induction of spinal anesthesia was achieved when at least the T6 dermatome was anesthetized; regression to T9 was considered as failure of anesthesia. Then the patients were returned to the lithotomy position after 3 minutes. Drug fixation time was 13 to 15 minutes (3 to 5 minutes for drug administration in the lateral position and 10 minutes for repositioning to the supine position and then lithotomy). If the blood pressure fell to 20% of the basic level, the patients received an intravenous bullous of 10 mg ephedrine. If the patients in the spinal group needed more than one injection of fentanyl at 2μg/kg, they would be excluded from the study, and ketamine and/or morphine were administered for relief of discomfort.

PCNL procedure

All procedures were performed with the patient in the prone position. Stone access was made by using fluoroscopic guidance, and the tract was dilated using a single-stage technique until 28 F to 30 F. Stones were extracted by grasper after breaking them by pneumatic lithotripter and holmium laser. A Double-J stent was not inserted in patients routinely, and nephrostomy tube insertion was optional and depended on surgeon preference.

Post-PCNL period

All patients received a solution including 1 mg/kg morphine in every 100 mL physiologic saline through the volumetric pump (patient controlled analgesia). This solution was infused at the rate of 2.5 mL/h during the 3-hour post-PCNL period in the recovery room; then morphine (0.05 mg/kg) was administered using intramuscular injection only according to the verbal rating scale greater than 3 after discharge from the recovery room until 24 hours after surgery. Serum creatinine level and blood cell count were measured 6 hours after surgery. On the first postoperative morning, radiography of the kidneys, ureters, and bladder (KUB) and renal ultrasonography were performed for all the patients to assess the status of stone clearance. Stone-free rate (SFR) was defined as residual stone ≤4 mm. KUB radiography and renal ultrasonography were reordered 1 month later for the patients with residual stone ≤4 mm.

Statistical analysis was performed by SPSS-19 (Statistical Package for the Social Sciences) software using independent paired t-test for quantitive data and chi-square test for qualitative variables.

Results

Mean age of patients in the general group and spinal group was 42.66±13.61 and 41.16±11.2 years (P=0.25), respectively. Distribution of stone location and stone burden revealed no significant difference between the two groups (P=0.66 and P=0.929). There were 98% of patients in the general group and 94% of patients in the spinal group who underwent PCNL using one access. Mean operative time (from caliceal puncture to nephrostomy tube insertion) was comparable in both groups (93.3±14.17 vs 89.8±10.51: P=0.89). Mean hospital stay in the general group and spinal group was 3.04 (2–5) and 2.82 (2–6) days, respectively (P=0.34) (Table 1).

Table 1.

Demographic, Preoperative, and Postoperative Data of Patients Who Underwent Percutaneous Nephrolithotomy According to Type of Anesthesia (General vs Spinal)

Variables	General anesthesia	Spinal anesthesia	P value
Age±SD (years)	42.66±13.61	41.16±11.2	0.25
Male/female (N/N)	27/23	29/21	0.07
Stone location	Pelvis (10)	Pelvis (7)	0.66
	Caliceal (13)	Caliceal (11)
	Pelvis and calix (24)	Pelvis and calix (30)
	Upper ureter (3)	Upper ureter (2)
Stone burden (cm²)	5.56±2.95	5.51±2.87	0.929
Access number (N)	One (49)	One (47)	0.364
	Two (1)	Two (3)
Operative time (min)	93.3±14.17	89.8±10.51	0.89
Preoperative hemoglobin (g/dL)	13.44±1.73	14.08±1.54	0.054
Hemoglobin level at discharge time (g/dL)	11.55±1.87	11.36±1.41	0.57
Preoperative GFR ( mL/min/1.73m²)	72.23±26.62	76.52±27.94	0.116
GFR at discharge time (mL/min/1.73m²)	68.45±24.65	74.89±27.12	0.04
Hospital stay (day)	3.04 (2–5)	2.82 (2–6)	0.34
Stone-free rate (%)	84%	79%	0.31

SD=standard deviation; GFR=glomerular filtration rate.

Preoperative hemoglobin level (13.44±1.73 in the general group vs 14.08±1.54 in the spinal group: P=0.054) and hemoglobin level at discharge time (11.55±1.87 in the general group vs 11.36±1.41 in the spinal group: P=0.57) were comparable between the two groups and, likewise, there was no significant difference between these patients in hemoglobin drop (preopative hemoglobin–hemoglobin level at discharge time) (P=0.18). Eight patients in the general group and 10 patients in the spinal group received packed red blood cells at the perioperative or postoperative period (P=0.6).

Mean preoperative glomerular filtration rate (GFR) (72.23±26.62 in the general group vs 76.52±27.94 in the spinal group: P=0.116) was insignificant on statistical analysis, but GFR at discharge time in the spinal group was better than the general group (68.45±24.65 in the general group vs 74.89±27.12 in table 1 the spinal group: P=0.04).

The SFR in the general group and spinal group was 84% and 79%, respectively, and there was no significant difference between the two groups regarding SFR (P=0.31). Mean residual stone after PCNL in the general group was 0.19 cm² and 0.32 cm² in the spinal group (P=0.33). Table 2 reveals detailed information about the necessity for auxiliary procedures after PCNL in patients with different types of anesthesia.

Table 2.

Requirement for Auxiliary Procedures After Percutaneous Nephrolithotomy in Patients with Different Types of Anesthesia

Auxiliary procedures	General anesthesia	Spinal anesthesia	P value
Re-PCNL^a	1	1	0.07
Ureteroscopy and Double-J stent insertion	4	2
Shockwave lithotripsy	4	9

Re-PCNL was performed in one patient in each group. These patients had complete staghorn renal calculi. We discussed with the patients regarding to the possible requirement for two sessions of percutaneous nephrolithotomy (PCNL); thus, this reoperation is not considered as a complication of surgery.

There were no complications compatible with grade IIIb, VI, and V in both groups. Sixteen patients in the general group and 13 patients in the spinal group had complications compatible with grade ≤II (P=0.76). Likewise, postspinal headache and considerable orthostatic hypotension were not observed in any patients in the spinal group. Distributions of different grades of complications according to the Clavien grading system are presented in Table 3.

Table 3.

Detailed Information About Perioperative and Postoperative Complications According to the Clavien System Grading

Complication	General anesthesia	Spinal anesthesia
Grade I	3	2
Grade II	13	11
Grade IIIa	4	2
Grade IIIb	0	0
Grade IV	0	0
Grade V	0	0

• Grade I: Fever less than 24 hours (two cases in general group and two cases in spinal group) – Nausea and vomiting after 6 hours post-PCNL that was controlled by anti-emetic agents (one case in general group).

• Grade II: Blood transfusion (eight in general group and ten in spinal group) – Fever >38.5C for more than 48 hours that was managed successfully using antibiotic therapy (five cases in general group and one case in spinal group).

• Grade IIIa: Four cases in general group and two cases in spinal group had urinary leakage after removing nephrostomy tube, so they underwent double-J stent insertion.

Postoperative vomiting was insignificant between the spinal (5%) and general (11%) groups (P=0.3). There was a significant difference between the general and spinal groups regarding analgesic requirement at the 24-hour post-PCNL period. Mean analgesic requirement in the spinal group was 6.8 mg and 13.2 mg in the general group (P<0.001) (Table 4).

Table 4.

Detailed Information About Mean Analgesic Requirement at Different Times After Percutaneous Nephrolithotomy

Study group	Mean morphine required within 6 hours after PCNL	Mean morphine required from 6 hours after PCNL until 24 hours	Mean morphine required within 24 hours after PCNL
Spinal	3.5 (3.8–7.1)	1.1 (0.5–1.6)	6.8 (4.1–9.6)
General	8.1 (6.4–12.8)	2.7 (0.7–4.8)	13.2 (10.7–17)
P value	P<0.001	P=0.135	P<0.001

Discussion

PCNL is the preferred treatment of patients with large, staghorn, and complex renal stones.¹³ In recent years, some efforts have been made to decrease morbidity and improve efficacy and surgical outcomes. Previous studies assessed some parameters that might influence the surgical outcomes such as use of different types of drainage (nephrostomy vs tubeless), access (supra- or infracostal) and position (flank, prone, supine, and prone-flexed),¹⁴ but the real impact of the type of anesthesia on PCNL outcomes has not been clarified precisely.

PCNL is performed in many centers using general anesthesia. Ballestrazzi and colleagues¹⁵ performed PCNL for the first time (1988) with regional anesthesia (epidural) in 112 patients and revealed acceptable findings in patient satisfaction and hemodynamic status, although they did not mention the effect of this type of anesthesia on surgical parameters. Saied and coworkers¹⁶ in 1991evaluated the role of regional anesthesia (spinal) in two small groups (25 patients in each group) by comparing the effect of bupivacaine in an interpleural block (IPB) after spinal anesthesia and combination of meperidine and diazepam. Their findings showed that surgery was totally pain free and the requirement for postoperative opium decreased significantly by using the IPB technique. In 2006, Atallah and associates¹⁷ compared the hemodynamic status and patient satisfaction in 108 patients who had undergone PCNL using bupivacaine and a combination of bupivacaine and fentanyl and revealed that spinal anesthesia is a reliable option for PCNL. The main drawback of the aforementioned studies was assessment of pain and anesthetic parameters and not surgical items.

Some of the previous studies have described their results in this manner by using CSEA or spinal anesthesia plus fentanyl as an alternative to general anesthesia. Singh and associates¹⁸ performed tubeless PCNL in 10 selected patients with spinal low-dose anesthesia with low-dose bupivacaine plus fentanyl. All patients were discharged the day after surgery (mean hospital stay was 40 hours), and no patients received blood transfusion and an analgesic agent. The authors believed that tubeless PCNL with regional anesthesia is a feasible technique and might decrease the convalescence period and analgesic requirement. Mehrabi and colleagues⁹ presented surgical outcomes after PCNL under spinal anesthesia using 15 mg bupivacaine plus 25 μg fentanyl in 160 patients. Mean operative time was 95 minutes, and the SFR was 71.8%. Ten (6.3%) patients received a blood transfusion, and 18 patients needed 10 mg ephedrine because of hypotension after spinal anesthesia. Even though they concluded that PCNL under spinal anesthesia can be a good alternative for general anesthesia, lack of a control group was the main defect of this study.

Kuzgunbay and associates⁷ compared different variables after PCNL using CSEA (15 mg bupivacaine and 25 μg fentanyl) in 37 patients and general anesthesia in 45 patients. Mean age, stone size, hemoglobin drop, SFR, and hospital stay were comparable between the two groups. CSEA was a feasible option for PCNL, especially in patients with high risk for general anesthesia and difficult intubation. We found only one randomized study¹² that had compared the surgical outcomes in 64 patients who had undergone PCNL with different types of anesthesia (CSEA vs general). Mean operative time, hospitalization period, fluoroscopic time, SFR, and blood transfusion rate were comparable between the two groups, but analgesic requirement in 24 hours after PCNL was significantly lower in the CSEA group (P<0.003).

Previous studies revealed that adding opioids (such as fentanyl) for dose reduction of the anesthetic agent (bupivacaine) may have adverse effects. There are some reports of respiratory arrest after opioid injection in regional anesthesia.¹⁹ Korhonen and colleagues²⁰ demonstrated that adding low-dose opium to an anesthetic agent in spinal anesthesia has a 75% chance of pruritus occurring. The overall incidence of pruritus after administration of 25 μg fentanyl as a protocol for CSEA was 90% in a randomized clinical trial study by Wells and coworkers.²¹ There were 28% to 45% of patients in different groups needed a rescue antipruritic agent such as naloxane. Recently, Pitkanen and coworkers²² revealed that the rate of fatality and neuraxial hematoma are significantly higher in epidural and CSEA than spinal anesthesia. The lateral holes of the epidural catheter may lie near to the dural puncture site; thus, injection of a bolus solution of drug through the epidural catheter has the potential risk of leakage from the dural puncture into the subarachnoid space.²³

Mean postoperative GFR at discharge was significantly better in the spinal group, but we have no clear and definitive hypothesis for this difference. Previous studies have not assessed the correlation between type of anesthesia and GFR changes. The effect of anesthetic agents on renal function or type and volume of fluid used during operation through intravenous or by irrigation may affect postoperative GFR changes.

PCNL is usually performed under general anesthesia with the patient in the prone position, and repositioning of the patient from supine to prone under an unawareness situation may lead to excessive head rotation and neck extension. Recently, Basiri and associates²⁴ presented unpredictable and devastating neurologic events after PCNL that happened in 11 patients who had undergone PCNL under general anesthesia in the prone position, but neurologic events were not observed in patients under spinal anesthesia. Perhaps, awareness of the patients under spinal anesthesia affected the decline in morbidity from extension of head and neck during repositioning.

This randomized study was designed to compare different surgical parameters and analgesic requirement after PCNL in patients with two types of anesthesia. Spinal anesthesia was performed by using bupivacaine (0.25 mg/kg), and no opium agents were used to reduce adverse side effects of opioids (respiratory arrest, severe pruritus). Operative time, hospital stay, SFR, hemoglobin drop, and complication rate were comparable between the two groups of patients, and analgesic requirement during 24 hours post-PCNL and mean GFR at discharge time were better in the spinal group.

Likewise, spinal anesthesia results in avoidance of tracheal intubation and possible unpredictable complications such as neurologic events, decrease in anesthesia controlled time, and more cost saving rather than general anesthesia. Thus, regarding comparable surgical outcomes of PCNL and reduction of postoperative analgesic requirement after spinal anesthesia, we think that PCNL under spinal anesthesia can be a considerable alternative to general anesthesia.

Conclusion

It seems that using spinal anesthesia solely by intrathecal injection of local anesthetic solutions instead of CSEA vs general anesthesia has comparable surgical outcomes and reduces the analgesic requirement and GFR drop after PCNL in the early postoperative period. Thus, PCNL using spinal anesthesia instead of general anesthesia with the patient in the prone position can be recommended as a first option in noncomplicated cases, but safety and efficacy of this approach in patients with a single functioning kidney, renal anomaly and/or skeletal deformity, history of PCNL or open stone surgery, and high risk for anesthesia (ASA score >3) has not been clarified and needs other well-designed studies.

Footnotes

Disclosure Statement

No competing financial interests exists.

Abbreviations Used

References

Gravenstein

. Extracorporeal shock wave lithotripsy and percutaneous nephrolithotomy. Anesthesiol Clin North America, 2000; 18:953–971.

Dasgupta

, Patel

. Percutaneous nephrolithotomy: Does position matter?—prone, supine and variations. Curr Opin Urol, 2013; 23:164–168.

Duty

, Waingankar

, Okhunov

et al. Anatomical variation between the prone, supine, and supine oblique positions on computed tomography: Implications for percutaneous nephrolithotomy access. Urology, 2012; 79:67–71.

Honey

, Wiesenthal

, Ghiculete

et al. Comparison of supracostal versus infracostal percutaneous nephrolithotomy using the novel prone-flexed patient position. J Endourol, 2011; 25:947–954.

Karami

, Rezaei

, Mohammadhosseini

et al. Ultrasonography-guided percutaneous nephrolithotomy in the flank position versus fluoroscopy-guided percutaneous nephrolithotomy in the prone position: A comparative study. J Endourol, 2010; 24:1357–1361.

Andreoni

, Olweny

, Portis

et al. Effect of single-dose subarachnoid spinal anesthesia on pain and recovery after unilateral percutaneous nephrolithotomy. J Endourol, 2002; 16:721–725.

Kuzgunbay

, Turunc

, Akin

et al. Percutaneous nephrolithotomy under general versus combined spinal-epidural anesthesia. J Endourol, 2009; 23:1835–1838.

Barak

, Putilov

, Meretyk

, Halachmi

. ETView tracheoscopic ventilation tube for surveillance after tube position in patients undergoing percutaneous nephrolithotomy. Br J Anaesth, 2010; 104:501–504.

Mehrabi

, Karimzadeh Shirazi

. Results and complications of spinal anesthesia in percutaneous nephrolithotomy. Urol J, 2010; 7:22–25.

10.

Borendal Wodlin

, Nilsson

, Carlsson

, Kjølhede

. Cost-effectiveness of general anesthesia vs spinal anesthesia in fast-track abdominal benign hysterectomy. Am J Obstet Gynecol, 2011; 205:326.el–7.

11.

McLain

, Kalfas

, Bell

et al. Comparison of spinal and general anesthesia in lumbar laminectomy surgery: A case-controlled analysis of 400 patients. J Neurosurg Spine, 2005; 2:17–22.

12.

Singh

, Sinha

, Sankhwar

, Malik

. A prospective randomized study comparing percutaneous nephrolithotomy under combined spinal-epidural anesthesia with percutaneous nephrolithotomy under general anesthesia. Urol Int, 2011; 87:293–298.

13.

Segura

, Preminger

, Assimos

et al. Nephrolithiasis Clinical Guidelines Panel summary report on the management of staghorn calculi. The American Urological Association Nephrolithiasis Clinical Guidelines Panel. J Urol, 1994; 151:1648–1651.

14.

Jun-Ou

, Lojanapiwat

. Supracostal access: Does it affect tubeless percutaneous nephrolithotomy efficacy and safety? Int Braz J Urol, 2010; 36:171–176.

15.

Ballestrazzi

, Zboralski

, Smith-Morel

et al.

[Importance of suspended peridural anesthesia in percutaneous nephrolithotomy. Apropos of 112 patients in the urology service of the Regional Hospital Center of Lille]

(Fre) Cah Anesthesiol, 1988; 36:85–88.

16.

Saied

, Sonbul

, el-Kenawy

, Atallah

. Spinal and interpleural bupivacaine for percutaneous nephrolithotomy. Middle East J Anesthesiol, 1991; 11:259–264.

17.

Atallah

, Shorrab

, Abdel Mageed

, Demian

. Low-dose bupivacaine spinal anaesthesia for percutaneous nephrolithotomy: The suitability and impact of adding intrathecal fentanyl. Acta Anaesthesiol Scand, 2006; 50:798–803.

18.

Singh

, Kumar

. “Ambulatory PCNL” (tubeless PCNL under regional anesthesia)—a preliminary report of 10 cases. Int Urol Nephrol, 2005; 37:35–37.

19.

Katsiris

, Williams

, Leighton

, Halpern

. Respiratory arrest following intrathecal injection of sufentanil and bupivacaine in a parturient. Can J Anaesth, 1998; 45:880–883.

20.

Korhonen

, Valanne

, Jokela

et al. Ondansetron does not prevent pruritus induced by low-dose intrathecal fentanyl. Acta Anaesthesiol Scand, 2003; 47:1292–1297.

21.

Wells

, Paech

, Evans

. Intrathecal fentanyl-induced pruritus during labour: The effect of prophylactic ondansetron. Int J Obstet Anesth, 2004; 13:35–39.

22.

Pitkänen

, Aromaa

, Cozanitis

, Förster

. Serious complications associated with spinal and epidural anaesthesia in Finland from 2000 to 2009. Acta Anaesthesiol Scand, 2013; 57:553–564.

23.

Vartis

, Collier

, Gatt

. Potential intrathecal leakage of solutions injected into the epidural space following combined spinal epidural anaesthesia. Anaesth Intensive Care, 1998; 26:256–261.

24.

Basiri

, Soltani

, Kamranmanesh

et al. Neurologic complications in percutaneous nephrolithotomy. Korean J Urol, 2013; 54:172–176.