A Comparative Study of Epidural Anesthesia with Dexmedetomidine Infusion versus General Anesthesia for Percutaneous Nephrolithotomy.

Background: There has been considerable debate regarding the ideal anaesthetic technique for Percutaneous Nephrolithotomy (PCNL). PCNL is usually performed under general anaesthesia (GA) in prone position. The prone position under GA is associated with various complications. To address these complications, our study was carried out to determine whether epidural anaesthesia [EA] with dexmedetomidine sedation can be a better alternative to GA for PCNL. Aims and
Objectives: To compare the efficacy and feasibility of performing PCNL under EA in combination with dexmedetomidine infusion. Methods and Material: Out of 225 patients observed in this study, 115 patients (group A) underwent PCNL under EA with dexmedetomidine infusion and 110 patients (group B) underwent PCNL under GA.
Results: Mean time to the first dose of rescue analgesia was significantly increased in Epidural group (328.17 ± 63.74) compared to GA group (72.09 min, p < .0001) and the mean Visual analog scale [VAS] scores were significantly lower in epidural compared to GA group at different time intervals during the first 24 h after surgery (p<0.05). Patients in epidural group had significantly less post-operative nausea, vomiting (6.1 % vs 13.6 %), and significantly less shivering (12.2% vs 33.6%). Mean patient satisfaction score in epidural group was significantly higher (8.75 ± 1.29 vs 8.14 ± 1.39, p=0.001); however, the mean surgeon satisfaction score was comparable among the two groups (8.76 ± 1.39 in Group A and 8.61 ± 1.35 in Group B, p=0.421). Conclusions: Our study shows that EA is an equally effective alternative to GA for PCNL, with more patient satisfaction, less postoperative pain, early ambulation and postoperative recovery, less systemic analgesic requirements and less adverse effects. Copyright:

INTRODUCTION

Percutaneous nephrolithotomy (PCNL) is the preferred treatment for large renal stones.[123] It facilitates a direct approach to the calculus while the kidney and surrounding structures are subjected to lesser trauma. It has been modified over time to reduce morbidity and hospital stay with the use of regional blocks, single step dilatation, “Mini-Perc” technique, tubeless PCNL, and sandwich therapy.[456]

There has been considerable debate regarding the ideal anesthetic technique for PCNL. The choice of anesthetic technique mainly depends on patient and surgeon preferences, feasibility of technique in a given patient, intra- and postoperative pain control, skills of anesthesiologist, and perioperative costs.[7]

PCNL is usually performed under general anesthesia (GA) in the prone position.[8] Postoperative vision loss (POVL) is a rare but devastating complication following prolonged procedure in the prone position under GA.[9] POVL mainly reported after spine surgery in the prone position has also been seen and reported after laparoscopic nephrectomies and robotic prostatectomies.[10] The change of position from supine to prone, under GA can cause complications such as displacement of the endotracheal tube, eye and brachial plexus injury, other neurological complications, and occasional spinal cord injury.[11] To address these complications, alternative anesthetic techniques such as neuraxial anesthesia: epidural, spinal and paravertebral have been tried.[1213]

Regional anesthesia (RA) for PCNL was first described in 1988.[14] Since then, a few studies have been done using RA for PCNL.[151617181920] RA is associated with lower morbidity and mortality than GA.[21] Associated complications and cost are higher for GA than for RA.[22] Besides, other complications related to PCNL such as pneumothorax, pleural effusion, and hemothorax can be picked up and intervened early under neuraxial anesthesia than under general anesthesia.[23] When RA is used, many of the issues related to positioning are resolved as the patients are conscious and can position themselves in the prone position according to their comfort.[24] RA has the advantage over GA in many urogenital surgeries. Because of the possibility of fluid absorption and electrolyte imbalance, RA may be a good alternative to GA.[25] We conducted a study in patients undergoing PCNL for nephrolithiasis comparing epidural anesthesia (EA) combined with dexmedetomidine infusion with GA to determine whether EA with dexmedetomidine sedation offers a suitable alternative to GA for PCNL.

SUBJECTS AND METHODS

After approval from the Institutional Ethical Committee (Ref number. IEC-SKIMS RP 57/2018.) and informed patient consent, this prospective study was conducted in the department of anesthesiology and department of urology and kidney transplant at (Sher-i-Kashmir Institute of Medical Sciences) over a period of 2 years. A total of 225 patients were randomly included in this study using random number tables. One hundred and fifteen patients in Group A underwent PCNL under EA and 110 patients in group B underwent PCNL under GA. The study included patients older than 18 years, with renal stones larger than 2 cm. Patients with skeletal deformity, renal anomaly, bleeding diathesis, anticoagulation use, the American Society of Anaesthesiologists (ASA) physical status >III were excluded from the study.

All patients in Group A received a bolus of 500 mL colloidbefore induction. Group A patients received EA at T9-T10/T10-T11 with 18 G Tuohy's needle and 20 G epidural catheter. All patients in Group A received 10–15 mL of 0.5% ropivacaine with 100 μg Fentanyl after a negative test dose of 3 mL of 1.5% lidocaine with adrenaline (1:200000) followed by Dexmedetomidine infusion at a rate of 0.2–0.7 μg.kg-1.h-1 intravenous (i.v.) and 1000 mg of i.v. paracetamol.

Group B patients received GA with propofol (1.5–2.5 mg.kg-1), fentanyl (2 μg.kg−1) followed by rocuronium (0.5 mg.kg-1) and 1000 mg of i.v. paracetamol. All the patients were mechanically ventilated and anesthesia was maintained with isoflurane (1%) in N2O and O2 mixture (50:50) and rocuronium (0.25 mg.kg-1) repeated as desired throughout the surgical procedure. Intraoperative hemodynamics, respiratory parameters, and adverse events were noted till the end of the surgery.

Time of administration of the first rescue analgesic (i.v. tramadol 0.5 mg.kg-1 in 10 mL saline given at the Visual Analog Scale [VAS] >3) was documented. Tramadol was used as morphine is not used on the floor. VAS scores were noted at 1, 3, 6, 12, 18, and 24 h postoperatively. Total analgesic requirement was also noted.

Patients were discharged on postoperative day 3. Surgeons were asked to opine regarding ease of patient positioning, locating pelvi-calyceal system, and overall comfort using a rating of good, fair or poor, immediately after surgery. Patients were asked to rate their overall satisfaction and overall comfort on a numerical scale (numeric rating scale 0–10) on postoperative day 2 in the surgical ward. Scores ranging from 0 to 2 were considered very poor, 3–5 poor, 6–8 fair and 9–10 good. Reasons for fair, poor, and very poor satisfaction were sought and noted.

RESULTS

The patients in both groups were comparable in terms of age, gender, body mass index, and ASA physical status [Table 1]. Mean baseline heart rate (beats. min-1) in Group A was 83.75 ± 5.98, whereas it was 80.42 ± 5.80 in Group B. The difference in the heart rate was statistically significant (P < 0.001). The mean heart rate in Group A at 5 min, 10 min, 20 min, and 40 min was lower than Group B. The difference in the heart rate was statistically significant (P < 0.05). At 60 min, 80 min, and 100 min, the mean heart rate in Group A was comparable to Group B [Table 2]. The mean arterial pressure (MAP) in Group A at 5 min, 10 min, 20 min, 40 min, and 60 min was higher as compared to Group B [Table 3]. Mean dexmedetomidine dose used in all the 115 patients of epidural group was 0.55 mg. kg-1.h-1. Mean time to the first rescue analgesic was 328.17 ± 63.74 min in Group A as compared to Group B where it was 72.09 ± 11.09 min. The difference between the two groups was statistically significant (P < 0.05) [Table 4].

Table 1

Demographic data

Group	Age (years), mean±SD	Gender		Mean±SD
		Male (%)	Female (%)	ASA status	BMI (kgm−2)
A (EA)	42.08±14.09	61.7	38.3	1.25±0.44	25.86±2.09
B (GA)	35.42±12.34	57.3	42.7	1.16±0.37	26.16±2.23
P	0.000*	0.137		0.103	0.305

*Statistically significant. Data expressed as mean±SD. EA=Epidural anaesthesia, GA=General anaesthesia, SD=Standard deviation, BMI=Body mass index, ASA=American society of anesthesiologists

Table 2

Comparison of mean heart rate

Variables (min)	Mean±SD		P
	Group A (EA)	Group B (GA)
HR 0	83.75±5.98	80.42±5.80	0.000*
HR 5	79.05±5.88	83.98±14.65	0.001*
HR 10	75.15±5.47	84.26±17.31	0.000*
HR 20	72.69±4.93	79.98±12.02	0.000*
HR 40	73.32±4.50	76.68±7.95	0.000*
HR 60	74.74±4.22	75.36±7.64	0.446
HR 80	76.20±4.31	75.34±8.10	0.316
HR 100	76.94±4.39	76.53±7.97	0.686
HR 120	78.55±4.82	74.19±9.26	0.001*

*Statistically significant. Data expressed as mean±SD. EA=Epidural anaesthesia, GA=General anaesthesia, SD=Standard deviation, HR=Herat rate

Table 3

Comparison of mean arterial pressure

MAP (min)	Mean±SD		P
	Group A (EA)	Group B (GA)
0	96.31±10.92	87.90±6.17	0.000*
5	88.04±7.95	81.64±5.47	0.000*
10	85.01±6.86	77.89±4.87	0.000*
20	83.74±5.20	80.53±4.65	0.000*
40	82.73±4.74	81.23±4.34	0.014
60	82.87±5.25	81.25±3.93	0.009*
80	84.09±4.70	83.54±3.98	0.350
100	84.82±5.11	83.90±4.39	0.150
120	84.81±4.99	84.11±5.05	0.417

*Statistically significant. Data expressed as mean±SD. EA=Epidural anaesthesia, GA=General anaesthesia, SD=Standard deviation

Table 4

Time to first rescue analgesic dose

	Mean±SD		P
	Group A (EA)	Group B (GA)
TTFRA (min)	328.17±63.74	72.09±11.01	0.000*

*Statistically significant. Data expressed as mean±SD. EA=Epidural anaesthesia, GA=General anaesthesia, SD=Standard deviation, TTFRA=Time to first rescue analgesic

Mean VAS scores in Group A were lower as compared to Group B at different time intervals during the first 24 h. The difference of VAS scores between the two groups at all the intervals of time was statistically significant (P < 0.05) [Table 5]. Postoperative nausea vomiting was seen in seven patients (6.1%) in Group A and 15 patients (13.6%) in Group B. The difference was statistically significant (P < 0.05). Shivering was seen in 14 patients (12.2%) in Group A as compared to 37 patients (33.6) in Group B. The difference was statistically significant (P < 0.05). Pneumothorax was seen in 1 (0.9%) out of 115 patients in Group A and none of the 110 patients in Group B [Table 6].

Table 5

Mean Visual Analogue Scale scores in the two groups

Mean VAS score (h)	Mean±SD		P
	Group A (EA)	Group B (GA)
1	0.03±0.26	2.95±0.85	0.000*
3	1.97±0.95	7.16±0.58	0.000*
6	5.23±1.74	5.61±0.66	0.036*
12	5.36±0.81	5.71±0.46	0.000*
18	3.58±0.63	3.97±0.75	0.000*
24	2.43±0.62	2.80±0.60	0.000*

*Statistically significant. Data expressed as mean±SD. EA=Epidural anaesthesia, GA=General anaesthesia, SD=Standard deviation, VAS=Visual Analogue Scale

Table 6

Complications/side effects in the two groups

Complication	Group A, n (%)	Group B, n (%)	P
PONV	7 (6.1)	15 (13.6)	<0.05*
Shivering	14 (12.2)	37 (33.6)	<0.05*
Nerve injuries	0	0	Not applicable
Pneumothorax	1 (0.9)	0	>0.05

*Statistically significant. Data expressed as n (%).Group A=Epidural anaesthesia, Group B=General anaesthesia, PONV=Postoperative nausea and vomiting

Mean patient satisfaction score (PSS) was 8.75 ± 1.29 in Group A, whereas in Group B, it was 8.14 ± 1.39. The difference between mean PSS in the two groups was statistically significant (P < 0.05). Mean surgeon satisfaction score was 8.76 ± 1.39 in Group A, whereas it was 8.61 ± 1.35 in Group B. The difference between mean PSS in the two groups was statistically not significant (P > 0.05) [Table 7].

Table 7

Comparison of patient and surgeon satisfaction scores among the two groups

Variables	Group A (n=115), n (%)	Group B (n=110), n (%)	P
Patient satisfaction score
Very poor	0	0	0.183
Poor	6 (5.2)	13 (11.8)
Fair	36 (31.3)	29 (26.3)
Good	73 (63.4)	68 (61.8)
Mean±SD	8.75±1.29	8.14±1.36	0.001*
Surgeon satisfaction score
Very poor	0	0	0.969
Poor	8 (6.9)	8 (7.2)
Fair	32 (27.8)	29 (26.3)
Good	75 (65.2)	73 (66.3)
Mean±SD	8.76±1.39	8.61±1.35	0.421

*Statistically significant. SD=Standard deviation

DISCUSSION

PCNL is the preferred treatment for large (>2 cm) renal stones.[3] GA is the traditional anesthetic choice for renal surgeries because of the uncomfortable body position required for the procedure. RA supplemented with good sedation has been advocated recently. [25] GA carries its own risks and complications such as stress response at induction, difficult airway, intraoperative awareness, difficult extubation, postoperative restlessness, and agitation. A higher incidence of side effects such as nausea and vomiting can make GA a very unpleasant experience.[26]

Prone position under GA is associated with more complications than the supine position. Accidental extubation and kinking of endotracheal tube can occur during positioning of the patient under GA. Pressure on the eyeballs in the prone position under GA may lead to POVL. Pressure on the pinna and breasts can cause pressure necrosis. Pressure points if not padded adequately can lead to peripheral nerve injuries. When regional anesthesia is used, the issues related to positioning are resolved as the patients are conscious and can position themselves in the prone position according to their comfort.

Taking advantage of the good hemodynamic profile characteristics of ropivacaine and excellent sedative properties of dexmedetomidine, we planned our study to compare GA and EA in patients undergoing PCNL. Out of the 225 patients, 115 received epidural anesthesia and 110 received GA.

There was a significant fall in heart rate from the baseline value in epidural group compared to GA group. Overall mean heart rate (beats. min-1) in EA group (76.71 ± 4.94) was lower than GA group (78.5 ± 10.07) [Figure 1]. In the study conducted by Parikh et al., baseline hemodynamics were comparable in both groups. Intergroup comparison with respect to Heart rate (HR) showed significant difference (P = 0.001) between the two groups from 0 min upto 120th min; with mean HR in Group GA being higher than segmental epidural anesthesia (SEA) group.[27] This was in agreement with our observation.

Figure 1

Mean heart rate at various time intervals

The mean MAP in epidural group at 5 min, 10 min, 20 min, 40 min, and 60 min was higher as compared to GA group and the difference was significant [Figure 2]. Hypotension was not seen in either group likely because of preloading with 500 mL colloid. Parikh et al. observed a statistically significant fall in MAP from its baseline value in SEA group, but the patients were clinically stable.[27] This was in concordance with our observation. Kumawat et al. on the otherhand observed that there was no statistically significant difference in mean heart rate and MAP at different time periods between the two groups.[28] This was in contradiction to our observation.

Figure 2

Comparison of mean arterial pressure between Group A and Group B. Group A. Epidural Anesthesia. Group B. General Anesthesia

Mean VAS scores in epidural group were lower as compared to GA group at different time intervals during the first 24 h. The difference of VAS scores between the two groups at all the intervals of time was statistically significant [Figure 3]. The lower VAS scores in epidural group could mainly be due to better analgesic effect of neuraxial anesthesia. Our findings were in agreement with Tangpaitoon et al. who observed that there was less early postoperative pain in epidural group as compared to GA group.[20]

Figure 3

Comparison of Visual Analog Scale score between Group A and Group B. Group A. Epidural Anesthesia. Group B. General Anesthesia

Mean time to the first rescue analgesic was 328.17 ± 63.74 min in Group A as compared to Group B where it was 72.09 ± 11.09 min. The difference between the two groups was statistically significant [Table 4]. Mean analgesic requirement within 24 h in Group B was 300 mg i.v. tramadol and in Group A it was 30.48±6.24 mg epidural ropivacaine. Parikh et al. found total tramadol requirements were significantly higher in GA group.[27]

Postoperative nausea vomiting was seen in 6.1% patients in epidural group and 13.6% patients in GA group [Table 6]. The difference was statistically significant. Higher incidence of postoperative nausea and vomiting in GA group may be due to the use of volatile anaesthetics and nitrous oxide and the administration of neostigmine and perioperative opioids which are known risk factors for PONV. Strategies to reduce risk include avoidance of GA by the use of RA, preferential use of propofol infusion, avoidance of nitrous oxide and volatile anaesthetics, minimizing perioperative opioids, and adequate hydration.[29] The consequences of PONV in the PACU include delayed discharge from the PACU, unanticipated hospital admission, increased incidence of pulmonary aspiration, and significant postoperative discomfort and patient dissatisfaction. Our results are in agreement with Tangpaitoon et al. who observed that 23.07% patients in GA Group and 4.19% in epidural group had postoperative nausea and vomiting.[20]

Postoperative shivering was seen in 12.2% patients in Group A as compared to 33.6% in Group B [Table 6]. The difference was statistically significant. The lower incidence of shivering in Group A patients can probably be attributed to anti-shivering properties of dexmedetomidine. Peri-operative administration of dexmedetomidine markedly diminishes the incidence of shivering in patients undergoing laparoscopic surgeries.[30]

Pneumothorax was seen in 0.9% patients in epidural group [Table 6]. This patient had a upper calyceal calculus. Pneumothorax was not seen in any patient in GA group. As the patient was awake under RA, she started complaining of difficulty in breathing, the pneumothorax was diagnosed early and was managed with chest drain insertion in the postanesthesia care unit. Had the patient been under GA, the pneumothorax might not have been detected this fast. Parikh et al. reported that one patient in GA group had pleural injury which was managed with an intercostal drain and was discharged on day 5.[27] Nerve injury was not reported in any of the patients in either group. Mean PSS was higher in epidural group [Table 7]. Greater patient satisfaction noted in epidural group may be attributed to better postoperative analgesia and mobility in the postoperative period and less PONV. Similar to our observation, Kumawat et al. observed that patients under regional EA had greater satisfaction with the anesthetic technique.[28] Mean surgeon satisfaction score was comparable [Table 7].

CONCLUSION

Thus, we conclude that PCNL under EA is as effective as PCNL under GA, with more patient satisfaction, less postoperative pain, early ambulation and postoperative recovery, and less adverse effects. EA requires skilled anesthetist and works well with expert and cooperative surgeons. It has the advantage of patient self-positioning, prolongation of anesthesia, safer in high-risk patients and provides better postoperative analgesia. Limitation to our study was that only ASA PS 1 and III patients were included and patients <18 year of age were excluded.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.