DOI: 10.31038/JCRM.2021411

Abstract

Background and objectives: Intra Operative Nausea and Vomiting (IONV), is distressing for patients, obstetricians, anesthetists; and may increase the risk of visceral injury during surgery by involuntary uncontrolled abdominal movements. In this study, we aimed to compare the antiemetic efficacy of low dose Ketamine versus Dexamethasone to decrease the incidence of IONV during Cesarean Section (C/S) under spinal anesthesia.

Materials and methods: This study was performed on 135 full term parturient women, aged between 18 and 40 years; candidate for cesarean section under spinal anesthesia at operation room. The parturient were allocated randomly to three groups using randomized blocking method. Group I (n = 45) received 8 mg Dexamethasone with total syringe volume of 5 ml, while group II (n = 45) received 20 mg Ketamine with total syringe volume of 5ml, and control group III (n = 45) received 5 ml normal saline. During the intraoperative period, the number of nausea, retching and vomiting episodes were recorded by an anesthetist who was blinded to the drug administered to the patient. Intraoperative hypotension and Bradycardia were recorded. The patient was also requested to report the symptoms of nausea, vomiting and shivering that may occur at the intervals. Vomiting was managed by Metoclopramide 10 mg slowly IV. A standardized surgical technique was used in all cesarean sections. Finally the statistical analysis was done using Statistical Package for Social Science version 21 program (SPSS21).

Results: This study, performed on 135 parturient women divided into three groups with 45 members in each one (first group received Dexamethasone, second and third groups were Ketamine and Placebo receivers in respect). Average age was 30.35 ± 5.94 years in Dexamethasone group, 29.97 ± 6.18 years in Ketamine group, and 29.6 ± 6.03 in control group(P = 0.840). In this study there was statistically insignificant increase in the rate of successful prevention of IONV (P = 0.062) and shivering (P = 0.550) using preoperative Ketamine and Dexamethasone. But there was statistically significant decrease in the rate of IONV in Ketamine group when compared to Dexamethasone group. The studied groups were comparable as regard to mean arterial blood pressure and heart rate, there was no statistically significant difference between groups during the study period. Also intraoperatively, there was no statistically significant difference in hypotensive episodes (P = 0.885), total Ephedrine administered (the average amount in those who suffered hypotension and received Ephedrine) (P = 0.623), and bradycardia (P = 0.146) between the studied groups. There was statistically significant decrease in the incidence of bradycardia in Ketamine group when compared to Dexamethasone group.

Conclusion: This study showed that there was no statistically significant decrease difference between groups receiving Dexamethasone and low dose Ketamine compared to control group during the operation period regarding the decrease rate of nausea and vomiting, hypotension and shivering.

Introduction

Minor side effects such as nausea and vomiting have been seen in more than 66% of cases [1-3]. Sudden contractions of Diaphragm during labor, which is unpleasant for patients can occur due to manipulation and abdominal traction. IONV (Intra Operative Nausea and Vomiting) is distressing for patients, obstetricians, anesthetists; and may increase the risk of visceral injury during surgery by involuntary uncontrolled abdominal movements [4-6]. Many factors may contribute to this high rate of IONV on the patients undergoing abdominal surgery under regional anesthesia: psychological changes due to anxiety, sympathetic block and the resultant hypotension secondary to spinal anesthesia, probably are the most important factors. Hypotension can induce the emetic symptoms by leading to cerebral hypo perfusion [7-9]. In addition, the chance of nausea and vomiting during labor is supposedly increased due to hormonal changes and increased internal pressure of uterus [10]. Considering the Causes of nausea and vomiting and its side effects; improvement of these conditions can have a significant effect on maternal treatment at the time of cesarean section. Prevention of hypotension or prescription of some kind of effective drugs is therefore important for the prevention of IONV. Dexamethasone, Ondansetron, ketamine, Pethidine and tramadol are some of the several drugs which are being used for prevention and treatment of the mentioned unwanted event [4,11-18]. Numerous studies have shown the effect of Dexamethasone in reducing nausea and vomiting [2,4,11]. It is assumed that Ketamine induced rising of blood pressure can reduce the incidence of spinal induced hypotension and consequently reduce nausea and vomiting [7,17,19]. In this study, we aimed to compare the efficacy of low dose ketamine versus Dexamethasone on IONV during C/S under spinal anesthesia.

Patients and Methods

This study is a type of randomized controlled, prospective, double blind clinical trial. The statistical population of the present study were patients referred to Alevi Educational and Medical Center of Ardabil city of Iran, with an indication for cesarean section during the year 2018. This study was performed on 135 ASA I–II female patients, between 18 and 40 years of age undergoing a planned C/S under spinal anesthesia and included three groups: Dexamethasone group, ketamine group and Placebo group, which were the recipients of sole normal saline. Sample volume is based on volume calculation sample for RCT studies with significance level of 0.05% and power of 80%; considering 60% = P1 in Placebo group and 28% = P2 in Dexamethasone group, 38 patients were counted for each group. In the same way, the sample volume is based on the volume calculation sample for RCT studies with significance level of 0.05 and power of 80%; and in terms of P1 = 60% in placebo group and P3 = 22% In the low dose ketamine group, 27 people for each group was calculated. And in terms of 20% possible loss, 7 people were added to Dexamethasone group and 18 patients were added to low dose ketamine group; thus 45 patients were enrolled in the present study to receive each drug for prevention of intraoperative nausea and vomiting.

After obtaining approval from the clinical research ethics committee of medical university with the code: IR.arums.rec.1396.230 and IRCT number: 20150808023559N18, all participants provided a written informed consent. 135 Patients were randomly divided into three groups using simple sampling method and block method (AABBCC) as follow: Ketamine group (including 45 patients receiving 20 mg Ketamine with total syringe volume of 5 ml), Dexamethasone group (including 45 patients receiving 8 mg Dexamethasone with total syringe volume of 5 ml) and control group (including 45 patients who received 5 ml of normal saline). All 3 syringes were similar in terms of color (content of all 3 syringes were colorless) and volume (all 3 syringes contained 5 ml of liquid substance). All three drugs were separately in syringes with specific codes of A,B,C on them. The three syringes were given by the second anesthetist to the anesthetist who was unaware of the content of the syringe and would administer it in a double-blind fashion before surgical incision and after spinal anesthesia. We excluded patients with, preeclampsia, eclampsia, psychiatric disorder, gastrointestinal disease, drug allergies, infection, diabetes, glaucoma, epileptic patients, and those on antiemetic agent in the last 24 hs. All parturient received 1000 ml Ringer solution IV over 30 min through 18 or 20 G IV lines before spinal anesthesia. Patients evaluated with standard monitoring including electrocardiogram, noninvasive arterial blood pressure measurement and pulse oximetry. Spinal anesthesia performed using a 25-gauge spinal needle in sitting position, through the L3-4, or L4-5 interspace.2.5 ml of 0.5% heavy Marcaine (Marcaine Spinal Heavy Ampule 0.5%, Astrazeneca) was administered to the subarachnoid space. Parturient were moved to supine position with operation bed turned 20 degree left lateral tilt to decrease the Aortocaval compression caused by the uterus and reducing hypotension after spinal anesthesia. The parturient were allocated randomly to three groups, ketamine group (n = 45) received 20 mg Ketamine with total syringe volume of 5 ml, Dexamethasone group (n = 45) received 8 mg Dexamethasone diluted in 5 ml normal saline and control group (n = 45) received 5 ml normal saline. the three syringes were given by the second anesthetist to the anesthetist who was unaware of the content of the syringe and would administer it in a double-blind fashion slowly IV over1 min before surgical incision. Face mask oxygenation used to all patients at a rate of 5 L/min, the level of sensory blockage was evaluated before the surgical incision. We excluded Patients with insufficient level of analgesia who were in need of general anesthesia. Estimated fluid deficits and maintenance requirements were replaced with Ringer’s solution IV, Intraoperative hypotension (MABP less than 20% of the basal reading) was managed by increasing the infusion rate of Ringer solution with injecting increments of 10 mg Ephedrine IV and if hypotension persisted another bolus Ephedrine was given. Bradycardia (HR < 50 beat/min) managed by Atropine (0.5 mg). After delivery of the baby, routine use of 10 units Oxytocin IV plus 20 units slow infusion if necessary were given to all parturient to enhance uterine contraction. We recorded the incidence of hypotension and increments of Ephedrine. During the intraoperative period, nausea, retching and vomiting episodes were recorded by an anesthetist who was blinded to the type of drug administered to the patient, questioning the patient in frequent intervals about the emetic symptoms. The patient was also requested to report any symptoms that may occur. Vomiting was managed by Metoclopramide 10 mg slowly IV, and IV analgesic (Fentanyl 50 microgram) was injected and the amount was recorded if required after delivery of the baby. A standardized surgical technique was used in all cesarean sections by Pfannenstiel incision. Nausea and Vomiting, evaluated through the following scoring table: Score 1: No nausea and vomiting, score 2: Only nausea, score 3: one to two episodes of nausea and vomiting, score4: Nausea and vomiting more than twice. Patients’ shivering score was evaluated according to the following shivering table: score0: No shivering observed, score 1: goose pumps and bristling, score 2: shivering localized to the group of body muscles, score 3: shivering in different bunches of body muscles score 4: shivering involves movements of the entire body.

The statistical analysis was done using SPSS program (statistical package for social science) version 21, Anova, Chi Square and Fisher Tests. All the tests were considered meaningful at the level of 0.05. All information and the answers were confidential.

Results

This study performed on 135 parturient women divided into three groups with 45 members in each one (first group received Dexamethasone, second and third groups were Ketamine and Placebo receivers in respect). There was no statistically significant differences in age between the three groups (P = 0.840). In the analysis of educational background, it was observed that 29 patients in Dexamethasone group (64.4%), 29 patients in ketamine group (64.4%) and 26 patients in control group (57.8%) were undergraduate. Anthropometric findings were observed and there was no significant difference in BMI among Dexamethasone, Ketamine and control groups (P = 0.246). ASA evaluated in 6 levels and it was observed that the majority of the patients in three groups were at I-ASA level (P = 0.165). All pregnancies were full term. None of the patients in the three groups had history of previous cardiac disease. None of them had history of cigarette smoking. Anesthetic level was adequate in all patients and sensory block was at T4 level. There was statistically insignificant increase in the rate of successful prevention of IONV in Ketamine group when compared to Dexamethasone group (P = 0.062). The studied groups were comparable as regard to systolic blood pressure, diastolic blood pressure, mean arterial blood pressure and heart rate, there was no statistically significant difference between groups at different spectrum of time during the study period (before surgical incision up to 75 minutes after operation). After analyzing the data in all three groups using Repeat measurement test regarding Systolic blood pressure there was significant difference between three groups. There were significant differences regarding diastolic blood pressure, mean arterial blood pressure and heart rate in each group during the different periods of time throughout the operation. The highest rate of bradycardia occurred in Dexamethasone group with 6 patients (13.3%) (P = 0.146). After two-by-two analyzing of the data in the study groups, it was observed that the incidence of bradycardia in Ketamine group was significantly lower than Dexamethasone group (P = 0.049). while there was no significant differences between Ketamine group compared with control group (P = 0.091), and Dexamethasone group compared with control group (P = 0.748).

Intraoperative hypotension was evaluated in three groups (MABP less than 20% of basic level) and the group with the most hypotensive cases was Dexamethasone group with 28 cases (62/2%); there were 26 patients (57/8%) in each of Ketamine and placebo groups. Analysis of results showed that this finding is not statistically significant.

Regarding shivering, it was observed that shivering occurred in 1 patient at Dexamethasone group with grade 1(2/2%) and 2patients in control group (1 patient in grade 1 and 1 patient in grade 2) (4/4%) and none of the patients in Ketamine group had shivering in three groups statistically was insignificant (P = 0.550).

The need for Ephedrine administration was analyzed and observed that in 32 patients of Dexamethasone group (71/1%), 31 patients of ketamine group (68/9%) and 32 patients of Control group (71/1%) single dose of Ephedrine was administered which was statistically insignificant.

There were no significant differences between three groups regarding the amount of received Ephedrine.

The average amount of received Ephedrine in Dexamethasone group was calculated 6.13 ± 16.22 mg, it was 18 ± 12/17 mg in ketamine group and 18/22 ± 48/12 mg in Control group (P = 0.623).

There was statistically significant increase in the rate of successful prevention of only nausea in Ketamine group when compared to Dexamethasone group (P = 0/023). The incidence of IONV was statistically insignificant when compared Ketamine group versus control group (P = 0/180) and Dexamethasone group versus control group (P = 0/335).

Regarding shivering, it was observed that there was single case of shivering with grade 1(2/2%) and 2 patients with grade 4(4/4%) in Dexamethasone group, and 1 patient with grade 1 and 1 patient with grade 2 in control group whereas no patient with shivering had been recorded in Ketamine group. There were statistically insignificant differences among three groups considering shivering (P = 0/550).

Discussion

In this study, 135 patients were divided in to three groups with 45 members in each (First group was recipient of Dexamethasone, second group was recipients of low dose Ketamine and the third group was placebo).

In the study performed by Ahmed Hassanein and Eissa Mahmoud [19] in Egypt in 2014 on 135 patients candidate for cesarean section, Aiming to determine the effects of low dose Ketamine versus Dexamethasone on intraoperative nausea and vomiting during C/S; The incidence of nausea and vomiting in Dexamethasone group (28/8%) and Ketamine group (22/2 %) (51/1%) were significantly lower than the control group.

But in the present study there was statistically insignificant difference in low dose ketamine group and dexamethasone group compared to the control group in terms of nausea and vomiting in the study performed by Kalani et al. [20] on 120 patients aiming the Effect of Ondansetron and Dexamethasone on nausea and Vomiting under Spinal Anesthesia, it was reported that in the 1st and 5th minutes after surgery, the incidence of nausea and vomiting, was 0% in ondansetron group and 10/2%, 13/3% respectively in Dexamethasone group, and it was observed that incidence of nausea and vomiting in ondansetron group was significantly lower.

In another study by Shalu et al. [21] titled “the Efficacy of Intravenous Dexamethasone in Prolonging the Duration of Spinal Anesthesia in Elective Cesarean Section” performed on 60 patients; Patients were divided into two groups.

The first group received Dexamethasone and the second group received Normal saline. It was observed that there were no statistically significant differences between the two groups in terms of nausea and vomiting which was identical to the findings of the present study.

In the study performed by Behdad et al. [22] on “Effects of Intravenous Ketamine during Spinal Anesthesia” in 60 Pregnant Women Undergoing Cesarean Section in 2013, the first group received 30 mg Ketamine with 1 mg Midazolam and the second group received Midazolam alone.

The statistical differences in nausea and vomiting in the first group Compared to the second group, was significantly high.

In the study performed by Modir et al. [23] under the title of “Prophylactic efficacy of Dexamethasone, Ketamine and Dexmedetomidine against intra- and postoperative nausea and vomiting under spinal anesthesia on 140 Pregnant woman undergoing cesarean section”; Patients were divided into 4 groups.

First group received normal saline, 2nd group received Dexamethasone, the 3rd group received Ketamine and the 4th group received Dexmedetomidine.

It was observed that all three drugs regarding nausea and vomiting are significantly effective compared to the control group and the incidence of nausea and vomiting is significantly less in Dexmedetomidine group compared to others.

In the present study the incidence of hypotension was insignificant in different groups (P = 0/885) while in the study of Hassanein et al. [19] there was no difference in Ephedrine administration (the average amount in those who suffered hypotension and received ephedrine), between Ketamine and dexamethasone groups, while the control group showed a significant high dose of total Ephedrine administration when compared to Ketamine and Dexamethasone groups (8 ± 3.2 versus 4.5 ± 2.1 in Ketamine group, and 5.3 ± 2 in Dexamethasone group), and a significant high rate of hypotensive episodes when compared to ketamine group (84.4% versus 64%).

In the study of Shalu et al. [21] heart rate, systolic blood pressure, diastolic blood pressure and arterial oxygen saturation in the Dexamethasone group had no significant difference with the control group; Which is in line with the results of our study. In the study of Kalani et al. [20] there were no significant differences between groups in mentioned items.

In Behdad et al. [22] study it was observed that Ketamine failed to make any significant changes to patients’ vital signs. In Modir et al. [23] the lowest heart rate (in normal range) and blood pressure was in Dexmedetomidine group and showed a significant difference compared to other three groups. But in our study, heart rate hadn’t any significant difference between Ketamine group and Dexamethasone group. In the study of Solhpour et al. [24] with the title of “ Comparison of prophylactic use of Meperidine, Meperidine plus Dexamethasone, and Ketamine plus Midazolam in preventing of shivering during spinal anesthesia” None of the four groups of this study showed any statistically significant difference in vital signs. Reviewing these studies, it was observed that Ketamine and Dexamethasone can’t make any significant changes in vital signs in a single dose and the majority of studies showed that vital signs were relatively more stable with these two drugs compared to the control group. In current study there wasn’t any significant difference regarding shivering among three groups. while in the study of Solhpour et al. [24] administration of combined Dexamethasone with Meperidine, could reduce the incidence of shivering. There was no significant differences among the four groups in nausea and vomiting. It also showed that Ketamin can reduce the incidence of nausea and vomiting while Dexamethasone can’t. The current study is in line with this finding.

Conclusion

This study showed that nausea and vomiting and bradycardia significantly decreased in low dose Ketamine group compared to Dexamethasone group while there weren’t any differences in hypotension and shivering between the two groups. There was no statistically significant difference between groups receiving Dexamethasone and low dose Ketamine compared to control group during the operation period regarding the decrease rate of nausea and vomiting, hypotension and shivering.

Acknowledgements

This article is derived from my final PhD thesis Which has been submitted to research deputy of Ardabil University of Medical Science and translated by me for the Journal of Clinical Research and Medicine. I would like to express my gratitude to my parents and my sisters who gave me the needed emotional support to persevere through what seemed like a totally overwhelming and never ending task.

References

1. Balki M, Carvalho JC (2005) Intraoperative nausea and vomiting during cesarean section under regional Int J Obstet Anesth 14(3): 230-241. [crossref]
2. Rodseth RN, Gopalan PD, Cassimjee  HM,  Goga  S  (2010)  Reduced  incidence of postoperative nausea and vomiting in black South Africans and its utility for a modified risk scoring Anesth Analg 110(6): 1591-1594. [crossref]
3. Griffiths JD, Gyte GML, Paranjothy S, Brown HC, Broughton HK, et al. (2012) Interventions for preventing nausea and vomiting in women undergoing regional anaesthesia for caesarean section. Cochrane Database Syst Rev 9: CD007579. [crossref]
4. Voigt M, Fröhlich CW, Hüttel C (2013) Prophylaxis of intra- and postoperative nausea and vomiting in patients during cesarean section in spinal Med Sci Monit 19: 993-1000. [crossref]
5. Ngan Kee WD, Khaw KS, Ng FF (2004) Comparison of phenylephrine infusion regimens for maintaining maternal blood pressure during spinal anesthesia for caesarean Br J Anaesth 92(4): 469-74. [crossref]
6. Cunningham FG, leveno KJ, Bloom SL (2010) Williams obstetrics and Gynecology 2010. 23th ed USA: McGraw-Hill, pp: 544-77.
7. Miller RD, Cohen NH, Eriksson LI, Fleisher LA, WienerKronish JP, Young Miller’s Anesthesia.8th ed. USA. Churchill Livingstone. 2015.
8. Apfel CC, Roewer N (2003) Risk assessment of postoperative nausea and Int Anesthesiol Clin 41: 13-32. [crossref]
9. Morgan EJ, Mikhail MS, Murray J (2006) Nonvolatile anesthetic agents, clinical anesthesiology. 4th , vol 1. Lange Medical Books/McGraw-Hill.
10. Rasooli S, Moslemi F, Khaki A (2014) Effect of subhypnotic doses of propofol and midazolam for nausea and vomiting during spinal anesthesia for cesarean section. Anesthesiol Pain Med 4(4): e19384. [crossref]
11. Kalani N, Zabetian H, Sanie MS, Deylami M, Radmehr M, et (2017) The Effect of Ondansetron and Dexamethasone on Nausea and Vomiting under Spinal Anesthesia. World Journal of Plastic Surgery 6(1): 88-93.
12. Pecora FST, Malbouisson LM, Torres MLA  (2009)  Supplemental  oxygen  and the incidence of perioperative nausea and vomiting in cesarean sections under subarachnoid Rev Bras Anestesiol 59: 558-569. [crossref]
13. Habib AS, Itchon-Ramos N, Phillips-Bute BG, Gan TJ (2006) Transcutaneous acupoint electrical stimulation with the Relief Band for the prevention of nausea and vomiting during and after cesarean delivery under spinal anesthesia. Anesth Analg 102: 581-584. [crossref]
14. Rowbotham DJ (2005) Recent advances in the nonpharmacological management of postoperative nausea and Br J Anaesth 95: 77-81. [crossref]
15. Demirhan A, Tekelioglu YU, Akkaya A, Ozlu T, Yildiz I, et al. (2013) Antiemetic effects of dexamethasone and ondansetron combination during cesarean sections under spinal African Health Sciences 13(2): 475-482. [crossref]
16. Voigt M, Fröhlich CW, Hüttel C, Kranke P, Mennen J, et al. (2013) Prophylaxis of intra- and postoperative nausea and vomiting in patients during cesarean section in spinal anesthesia. Medical Science Monitor: International Medical Journal of Experimental and Clinical Research 19: 993-1000. [crossref]
17. Rahmanian M, Leysi M, Hemmati AA, Mirmohammadkhani M (2015) The Effect of Low-Dose Intravenous Ketamine on Postoperative Pain Following Cesarean Section with Spinal Anesthesia: A Randomized Clinical Oman Medical Journal 30(1): 11-16. [crossref]
18. Behdad S, Hajiesmaeili MR, Abbasi HR, Ayatollahi V, Khadiv Z, et al. (2013) Analgesic Effects of Intravenous Ketamine during Spinal Anesthesia in Pregnant Women Undergone Caesarean Section; A Randomized Clinical Anesthesiology and Pain Medicine 3(2): 230-233. [crossref]
19. Hassanein A, Mahmoud E (2015) Effect of low dose ketamine versus dexamethasone on intraoperative nausea and vomiting during cesarean section under spinal anesthesia. Egyptian Journal of Anaesthesia 31: 59-63.
20. Kalani N, Zabetian H, Sanie MS, Deylami M, Radmehr M, et (2017) The Effect of Ondansetron and Dexamethasone on Nausea and Vomiting under Spinal Anesthesia. World J Plast Surg 6(1): 88-93. [crossref]
21. Shalu PS, Ghodki PS (2017) To Study the Efficacy of Intravenous Dexamethasone in Prolonging the Duration of Spinal Anesthesia in Elective Cesarean Section. Anesth Essays Res 11(2): 321-325. [crossref]
22. Behdad S, Hajiesmaeili MR, Abbasi HR, Ayatollahi V, Khadiv Z, et al. (2013) Analgesic Effects of Intravenous Ketamine during Spinal Anesthesia in Pregnant Women Undergone Caesarean Section; A Randomized Clinical Anesth Pain Med 3(2): 230-3. [crossref]
23. Modir H, Moshiri E, Kamali A, Shokrpour M, Shams N (2019) Prophylatic efficacy of dexamethasone, ketamine and dexmedetomidine against intra- and postoperative nausea and vomiting under spinal Formos J Surg 52: 17-23.
24. Solhpour A, Jafari A, Hashemi M, Hosseini B, Razavi S, et (2016) A comparison of prophylactic use of meperidine, meperidine plus dexamethasone, and ketamine plus midazolam for preventing of shivering during spinal anesthesia: a randomized, double-blind, placebo-controlled study. J Clin Anesth 34: 128-35. [crossref]

DOI: 10.31038/JCRM.2020345

Abstract

Objective: This study compared the perioperative outcomes from patients who underwent pancreaticojejunostomy in pancreaticoduodenectomy via the modified Kakita method anastomosis (KMA) or the modified Cattell-Warren anastomosis (CWA).

Summary of background data: We retrospectively evaluated 43 consecutive patients who underwent pancreaticoduodenectomy between January 2006 and December 2012.

Methods: The modified CWA was exclusively performed before December 2009, and the modified KMA was exclusively performed after January 2010. To evaluate their simplicity and safety, we compared the perioperative outcomes for the patients who underwent CWA (n = 22) and the patients who underwent KMA (n = 21).

Results: Pancreatic fistula was significantly less frequent in the KMA group, compared to in the CWA group (4.8% vs. 36.3% respectively, p = 0.021). In addition, the rate of all surgical complications decreased after the introduction of KMA at our institution.

Conclusions: The results of this retrospective study appear to indicate that KMA is a simpler and safer technique, compared to CWA, for pancreaticojejunostomy in pancreaticoduodenectomy.

Keywords

Kakita method anastomosis, Pancreaticoduodenectomy, Pancreaticojejunostomy, Pancreatic fistula

Manuscript Summary

The major finding(s) from the study: Our results indicate that the modified Kakita technique provided a significantly lower frequency of pancreatic fistula, as well as non-significant reductions in other postoperative complications.

What the findings add to existing knowledge: We conclude that the modified Kakita technique may be simpler and more effective than the modified Cattel-Warren technique.

What is already known in the field: U-sutures may reduce shear forces at the fragile pancreatic parenchyma, and subsequently reduce the incidence of pancreatic fistula.

What should change as a result: We will perform the modified Kakita method anastomosis.

Introduction

The history of pancreaticojejunostomy in pancreaticoduodenectomy has been described throughout the literature, with Whipple et al. reporting the first cases of pancreaticoduodenectomy in 1935 [1]. Whipple also introduced pancreaticojejunostomy with complete one-stage reconstruction in 1946 [2]. However, in 1943, Cattell stated that pancreaticoenteric anastomosis was indispensable, and maintained that leakage of the pancreatic juice accounted for many postoperative complications and deaths among patients who underwent pancreaticoduodenectomy [3]. Therefore, Cattell recommended direct anastomosis of the pancreatic duct and jejunum in patients with a main pancreatic duct that had a sufficient diameter. However, for smaller pancreatic ducts, Cattell recommended the use of a “necrosing suture”, whereby the pancreatic duct was ligated and the cut surface of the pancreas was covered with the jejunal wall. Unfortunately, pancreaticoduodenectomy has historically had high rates of complications and operative mortality, which were often related to suture failure during pancreaticojejunostomy in pancreaticoduodenectomy. However, some high-volume institutions have reported mortality rates of <5% for pancreaticoduodenectomy [4-7], although the postoperative morbidity rates remain high, ranging from 30% to 50% [6-13].

Pancreatic fistula is a well-known complication of pancreaticoduodenectomy, with rates of 2–20% being recently reported [7-17]. To address this issue, several different anastomotic techniques have been used to minimize the incidence of pancreatic fistula [15,16]. These techniques include the modified Cattell-Warren anastomosis (CWA) 3, Peng’s method [16], Blumgart’s method [17], invaginating the pancreatic stump into the jejunal stump [18], and the modified Kakita method anastomosis (KMA) [19,20]. In this retrospective study, we compared the perioperative outcomes for pancreaticoduodenectomy among patients who underwent the CWA and KMA procedures.

Methods

Patients

Between January 2006 and December 2012, 43 consecutive patients underwent pancreaticoduodenectomy with pancreaticojejunostomy in the Department of Gastroenterological Surgery at Tomei Atugi Hospital, and were entered into our prospective database. The modified CWA method was exclusively performed before December 2009, and the modified KMA method was exclusively performed after January 2010. Using a before-after cohort design, we compared the perioperative outcomes for the CWA (n = 22) and the KMA (n = 21) groups.

Surgical Technique and Postoperative Management

All surgical procedures were performed by or under the supervision of experienced pancreatic surgeons. Most patients underwent subtotal stomach-preserving pancreaticoduodenectomy (SSpPD), which involves resection of the pyloric ring and preservation of >95% of the stomach, although some patients underwent conventional pancreaticoduodenectomy with distal gastrectomy. Reconstruction was performed using a modified Child’s technique for both SSpPD and conventional pancreaticoduodenectomy. The anastomosis was performed (in order of preference) between the jejunum and pancreas, bile duct, and stomach. Drain tubes (8-mm silicone tubes) ware placed at the ventral and dorsal sides of the pancreaticojejunostomy. Oral fluids were started at 72 h after the surgery, and oral intake was started at approximately 5 days after surgery, except in cases with postoperative complications, such as delayed gastric emptying. All abdominal drains were removed at day 7 after the surgery if the drainage fluid was clear, did not exceed 300 mL per 24 h, and contained a concentration of amylase that was <3-fold greater than the serum concentration. Second-generation cephem antibiotics were administered immediately before surgery and every 3 h during surgery, with continuation until day 3 after the surgery. In cases that contracted an infectious disease, the antibiotics were changes as necessary; octreotide was not routinely used.

The Modified KMA Technique

The pancreatic duct and jejunal mucosa were joined in an end-to-side fashion, using eight absorbable interrupted sutures (PDSII 5/0, ETHICON) via the duct-to-mucosa anastomosis. All patients who underwent KMA had a 4-Fr to 6-Fr polyvinyl catheter inserted into the main pancreatic duct for external drainage. The unique aspect of this modified KMA technique is the approximation of the pancreatic parenchyma to the jejunal seromuscular layer, using five or six non-absorbable interrupted penetrating sutures (Prolene 3/0, ETHICON) [19,20].

The Modified CWA Technique

The modified CWA was performed after a small incision was made at the antimesenteric side of the jejunal loop. Monofilament absorbable interrupted sutures (PDSII 3/0, ETHICON) were placed using an atraumatic needle, beginning at the posterior surface of the pancreas. The dorsal capsule of the pancreas was sutured to the seromuscular layer of the jejunum, and then the central portion of the anastomosis was completed as a duct-to-mucosa anastomosis, using interrupted sutures (PDSII 5/0, ETHICON). Finally, monofilament absorbable interrupted sutures (PDSII 3/0, ETHICON) were placed at the anterior surface of the pancreas [3]. All patients who underwent CWA had a 4-Fr to 6-Fr polyvinyl catheter inserted into the main pancreatic duct for external drainage.

Data Collection and Evaluation Parameters

We retrospectively reviewed our institution’s database to obtain the following case-specific information: age, sex, preoperative biliary drainage, diagnosis, medical history, preoperative laboratory findings (serum glutamic oxaloacetic transaminase, bilirubin, alkaline phosphatase, albumin, creatinine, lipase, amylase, hemoglobin, white cell count, C-reactive protein, and partial thromboplastin time), body mass index, pancreatic texture, operative time, intraoperative blood loss, number and type of postoperative local and systemic complications, and mortality. Postoperative morbidity was defined as any postoperative surgical or non-surgical complication. Postoperative pancreatic fistula (POPF) was diagnosed and graded based on the International Study Group on Pancreatic Fistula guidelines. The all-inclusive definition was a drain output of any measurable fluid volume on or after postoperative day 3, with amylase concentration of >3-fold higher than the serum amylase concentration. Three different grades of POPF (grades A, B, C) were defined according to the clinical signs of infection and/or a necessary change in the clinical management [21]. A fistula of grade B (fistula requiring any therapeutic intervention) or higher was considered clinically significant.

Statistical Analysis

Consecutive data were expressed as median (range) and were analyzed using the Mann-Whitney U test. Inter-group differences in numerical data were evaluated using the χ2 test (with Yates correction) or Fisher’s exact test when the n-value was <5. All statistical analyses were performed using Ystat2013 (Microsoft Excel), and differences with a p-value of <0.05 were considered statistically significant.

Results

Patient Characteristics

This study evaluated 22 patients who underwent CWA and 21 patients who underwent KMA; their characteristics are shown in Table 1. However, there were no significant differences in age or sex when we compared the two groups. In the CWA group, the pathological diagnoses were pancreatic cancer in 10 patients, cholangiocarcinoma in 9 patients, and intraductal papillary mucinous neoplasms in 3 patients. In the KMA group, the pathological diagnoses were pancreatic cancer in 10 patients, cholangiocarcinoma in 9 patients, and cystic intraductal papillary mucinous neoplasm in 2 patients. When we compared the two groups, no significant differences were observed for pancreatic texture (hard/soft), mean operative time, or intraoperative blood loss.

Table 1: Patient characteristics.

IPMN: intraductal papillary mucinous neoplasm

Postoperative Complications

The types and frequencies of the postoperative complications are shown in Table 2. Pancreatic fistula occurred significantly less frequently in the KMA group, compared to in the CWA group (4.8% vs. 36.3%, p = 0.021), and one case of pancreatic fistula-related hemorrhage was observed in the CWA group. When we compared the specific incidences of pancreatic fistulas, grade B or C fistula was recognized in one case for the KMA group, compared to 7 cases for the CWA group, with latent presentation of a pancreatic fistula in one case. In the case with latent presentation of the pancreatic fistula, the drainage fluid amylase concentration was not elevated during the postoperative period, although the fistula was diagnosed via computed tomography after the drain was removed (Table 3). In addition, we observed a noticeable, although not significant, difference in the frequency of surgical complications after the introduction of KMA (23.8% after KMA vs. 45% after CWA; p = 0.242). Furthermore, the KMA group experienced fewer morbidities, although this difference was also not statistically significant (52.3% vs. 68.1%, p = 0.597). No cases of in-hospital mortality were observed for either group.

Table 2: Postoperative complication.

Table 3: Comparison of pancreatic fistula incidence for Cattell-Warren and Kakita method anastomosis.

Comparing the Drainage Fluid Amylase Concentrations and Duration of Drain Insertion

When we compared the two groups, no significant differences were observed in the median drainage fluid amylase concentration in the CWA and KMA groups (CWA: 98 IU/L; range, 2–83,900 IU/L; KMA: 45 IU/L; range, 6–1,036 IU/L) (Figure 1). The drainage fluid amylase concentration exceeded 1,000 IU/L in 4 cases (3 cases in the CWA group and one case in the KMA group) on or after postoperative day 3. When we compared the duration of drain insertion for both groups, no significant difference in the median duration was observed (CWA: 16 days; range, 7–94 days; KMA: 14 days; range, 7–57 days) (Figure 2).

Figure 1: Amylase concentrations in the drainage fluid for all cases.

No significant difference was observed when we compared the median amylase concentrations in the drainage fluids from the Cattell-Warren anastomosis (CWA) group (median, 98 IU/L; range, 2–83,900 IU/L) and the Kakita method anastomosis (KMA) group (median, 45 IU/L; range, 6–1,036 IU/L). P=0.088 via the Mann-Whitney U-test.

Figure 2: Duration of drain insertion for all cases.

No significant difference was observed when we compared the median duration of drain insertion for the Cattell-Warren anastomosis (CWA) group (median, 16 days; range, 7–94 days) and the Kakita method anastomosis (KMA) group (median, 14 days; range, 7–57 days). P=0.501 via the Mann-Whitney U-test.

Discussion

The techniques that are used for reconstruction of the pancreatic stump after pancreaticoduodenectomy are closely related to the incidence of postoperative complications, mortality, and reduced quality of life. Pancreatic fistula is a well-known complication of pancreaticoduodenectomy, with rates of 2–20% being recently reported [7,8,10,12,14-17,20,22,23]. In many institutions, several different surgical procedures, such as Blumgart anastomosis, have been used to minimize the incidence of pancreatic fistula. Among these procedures, CWA is the most well-known procedure, and has been commonly used for a long period of time. In contrast, KMA is a relatively simple technique, and many surgeons in Japan perform KMA in pancreas-jejunum anastomosis. In this retrospective study, we found that KMA appeared to be a simpler and safer technique for pancreaticojejunostomy, compared to CWA. Moreover, the KMA technique significantly reduced the frequency of pancreatic fistula, with non-statistically significant reductions for other postoperative complications.

Patient age and intraoperative blood loss have been identified as perioperative risk factors for pancreatic fistula. In addition, soft pancreatic texture, pancreatic duct size, and pancreatic juice output have been reported to be predictive factors for pancreatic fistula [24,25]. In the present study, we observed similar trends within both groups, although there were no significant differences when we compared the risk and predictive factors between the two groups.

In CWA, multiple sutures are placed tangentially through the pancreatic capsule, which may create shear forces at the fragile pancreatic parenchyma. Furthermore, the knot-tying may cause the sutures to cut through the pancreas, and the use of multiple sutures is known to cause pancreatic microleakage during the knot-tying [17]. Therefore, it has been speculated that the use of too many sutures and/or too aggressive knot-tying may cause ischemia and necrosis of the pancreatic stump. In contrast, KMA uses only five or six non-absorbable interrupted penetrating sutures to approximate the pancreatic parenchyma to the jejunal seromuscular layer. Thus, this technique reduces the total number of sutures, avoids placing unnecessary shear forces on the fragile pancreatic parenchyma, and avoids some of the complicated manipulations that are required for other surgical techniques [19]. Furthermore, the KMA technique can help to reduce the risk of suture failure as a result of necrosis and ischemia.

Various previous studies have compared different anastomosis techniques, such as pancreaticojejunostomy versus pancreaticogastrostomy [25], Blumgart anastomosis versus modified CWA [17] or versus the Kakita type anastomosis [26], pancreaticojejunostomy with the invagination technique (dunking) versus duct-to-mucosa pancreaticojejunostomy [27], or binding anastomosis [28]. In addition, prospective randomized trials have compared pancreaticojejunostomy to pancreaticogastrostomy, and found that both procedures provided similar incidences of pancreatic fistula [25,29,30]. Similarly, a comparison of the invagination method and pancreatic duct jejunum anastomosis found no difference in the incidence of pancreatic fistula [6]. However, Blumgart anastomosis was associated with a lower incidence of pancreatic fistula, compared to the modified CWA (4% vs. 13%, respectively) [17] or to Kakita type anastomosis (2.5% vs. 36%, respectively) [26]. Similarly, the recessed method has been reported to provide a low incidence of pancreatic fistula [27], and Peng et al. have reported pancreatic leakage rates of 0% using a complex three-layer dunking anastomosis [29,31,32], although this procedure is technically difficult. Interestingly, the Blumgart and “dunking” invagination techniques use U-sutures [33,34], and these techniques provide relatively low complication rates. Therefore, U-sutures may reduce shear forces at the fragile pancreatic parenchyma, and subsequently reduce the incidence of pancreatic fistula. Similarly, the KMA method attempts to reduce the shear force in a manner that is similar to that performed with U-sutures.

Unfortunately, despite various techniques having been developed to manage the pancreatic remnant after pancreaticoduodenectomy, none of these techniques are associated with clearly superior outcomes. Thus, it is important to preserve the pancreatic capsule and to avoid bleeding from the pancreatic parenchyma during pancreaticojejunostomy, which can affect hemostasis in that tissue. Therefore, it is important to use surgical and suturing techniques that preserves as much of the parenchyma as possible (by not placing unnecessary shearing force on the pancreas).

This study has several limitations. First, because it is a retrospective single-center study, there are limitations regarding the generalizability of our data. In addition, over the course of 6 years, there is a possibility that the postoperative management may have changed slightly. Furthermore, it is impossible to completely exclude the potential effect of confounders (e.g., surgical standards and perioperative management), although it is unlikely that these factors strongly influenced the incidence of pancreatic fistula and suture insufficiency. Nevertheless, our results indicate that KMA was a simple and safe technique for reducing the incidence of pancreatic fistula and leakage rates after pancreaticojejunostomy.

References

1. Whipple AO, Parsons WB, Mullins CR (1935) TREATMENT OF CARCINOMA OF THE AMPULLA OF VATER. Ann Surg 102(4): 763-779 [crossref]
2. Whipple AO (1946) Observations on radical surgery for lesions of the pancreas. Surg Gynecol Obstet 82: 623-631 [crossref]
3. Cattell RB (1949) The surgical treatment of carcinoma of the pancreatoduodenal area. Ann R Coll Surg Engl 4(4): 197-205 [crossref]
4. Poon RT, Lo SH, Fong D, Fan ST, Wong J (2002) Prevention of pancreatic anastomotic leakage after pancreaticoduodenectomy. Am J Surg 183(1): 42-52 [crossref]
5. Shapiro TM (1975) Adenocarcinoma of the pancreas: a statistical analysis of biliary bypass vs Whipple resection in good risk patients. Ann Surg 182(6): 715-721 [crossref]
6. Bassi C, Falconi M, Molinari E, Mantovani W, Butturini G. et al. (2003) Duct-to-mucosa versus end-to-side pancreaticojejunostomy reconstruction after pancreaticoduodenectomy: results of a prospective randomized trial. Surgery 134(5): 766-771 [crossref]
7. Buchler MW, Friess H, Wagner M, Kulli C, Wagener V. et al. (2000) Pancreatic fistula after pancreatic head resection. Br J Surg 87(7): 883-889 [crossref]
8. Takano S, Ito Y, Watanabe Y, Yokoyama T, Kubota N. et al. (2000) Pancreaticojejunostomy versus pancreaticogastrostomy in reconstruction following pancreaticoduodenectomy. Br J Surg 87(4): 423-427 [crossref]
9. Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL (2002) et al. Hospital volume and surgical mortality in the United States. N Engl J Med 346(15): 1128-1137 [crossref]
10. Yeo CJ, Cameron JL, Lillemoe KD, Sauter PK, Coleman J. et al. (2000) Does prophylactic octreotide decrease the rates of pancreatic fistula and other complications after pancreaticoduodenectomy? Results of a prospective randomized placebo-controlled trial. Ann Surg 232(3): 419-429 [crossref]
11. Grobmyer SR, Rivadeneira DE, Goodman CA, Mackrell P, Lieberman MD (2000) Pancreatic anastomotic failure after pancreaticoduodenectomy. Am J Surg 180(2): 117-120 [crossref]
12. Ohwada S, Ogawa T, Kawate S, Tanahashi Y, Iwazaki S. et al. (2001) Results of duct-to-mucosa pancreaticojejunostomy for pancreaticoduodenectomy Billroth I type reconstruction in 100 consecutive patients. J Am Coll Surg 193(1): 29-35 [crossref]
13. Shyr YM, Su CH, Wu CW, Lui WY (2003) Does drainage fluid amylase reflect pancreatic leakage after pancreaticoduodenectomy? World J Surg 27(5): 606-610 [crossref]
14. Strasberg SM, Drebin JA, Mokadam NA, Green DW, Jones KL, et al. (2002) Prospective trial of a blood supply-based technique of pancreaticojejunostomy: effect on anastomotic failure in the Whipple procedure. J Am Coll Surg 194(6): 746-758; discussion 759-760 [crossref]
15. Rault A, SaCunha A, Klopfenstein D, Larroude D, Epoy FN, et al. (2005) Pancreaticojejunal anastomosis is preferable to pancreaticogastrostomy after pancreaticoduodenectomy for longterm outcomes of pancreatic exocrine function. J Am Coll Surg 201(2): 239-244 [crossref]
16. Peng SY, Wang JW, Li JT, Mou YP, Liu YB, et al. (2004) Binding pancreaticojejunostomy–a safe and reliable anastomosis procedure. HPB (Oxford) 6(3): 154-160 [crossref]
17. Kleespies A, Rentsch M, Seeliger H, Albertsmeier M, Jauch KW. et al. (2009) Blumgart anastomosis for pancreaticojejunostomy minimizes severe complications after pancreatic head resection. Br J Surg 96(7): 741-750 [crossref]
18. Bugiantella W, Rondelli F, Mariani L, Longaroni M, Federici MT, et al. (2014) Pancreatico-jejunal anastomosis with invaginating jenunal “J”-loop. Preliminary report of a new technique. Int J Surg 12 Suppl 1: S87-90
19. Kakita A, Yoshida M, Takahashi T (2001) History of pancreaticojejunostomy in pancreaticoduodenectomy: development of a more reliable anastomosis technique. J Hepatobiliary Pancreat Surg 8(3): 230-237
20. Kakita A, Takahashi T, Yoshida M, Furuta K (1996) A simpler and more reliable technique of pancreatojejunal anastomosis. Surg Today 26(7): 532-535 [crossref]
21. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, et al. (2005) Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 138(1): 8-13 [crossref]
22. Kanda M, Fujii T, Kodera Y, Nagai S, Takeda S, et al. (2011) Nutritional predictors of postoperative outcome in pancreatic cancer. Br J Surg 98(2): 268-274 [crossref]
23. Lai EC, Lau SH, Lau WY (2009) Measures to prevent pancreatic fistula after pancreatoduodenectomy: a comprehensive review. Arch Surg 144(11): 1074-1080 [crossref]
24. Kollmar O, Moussavian MR, Bolli M, Richter S, Schilling MK (2007) Pancreatojejunal leakage after pancreas head resection: anatomic and surgeon-related factors. J Gastrointest Surg 11(12): 1699-1703 [crossref]
25. Bassi C, Falconi M, Molinari E, Salvia R, Butturini G, et al. (2005) Reconstruction by pancreaticojejunostomy versus pancreaticogastrostomy following pancreatectomy: results of a comparative study. Ann Surg 242(6): 767-771, discussion 771-763 [crossref]
26. Fujii T, Sugimoto H, Yamada S, Kanda M, Suenaga M et al. (2014) Modified Blumgart anastomosis for pancreaticojejunostomy: technical improvement in matched historical control study. J Gastrointest Surg 18(6): 1108-1115 [crossref]
27. Berger AC, Howard TJ, Kennedy EP, Sauter PK, Bower-Cherry M, et al. (2009) Does type of pancreaticojejunostomy after pancreaticoduodenectomy decrease rate of pancreatic fistula? A randomized, prospective, dual-institution trial. J Am Coll Surg 208(5): 738-747; discussion 747-739 [crossref]
28. Peng S, Mou Y, Cai X, Peng C (2002) Binding pancreaticojejunostomy is a new technique to minimize leakage. Am J Surg 183(3): 283-285 [crossref]
29. Yeo CJ, Cameron JL, Maher MM, Sauter PK, Zahurak ML, et al. (1995) A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy. Ann Surg 222(4): 580-588; discussion 588-592 [crossref]
30. Duffas JP, Suc B, Msika S, Fourtanier G, Muscari F, et al. (2005) A controlled randomized multicenter trial of pancreatogastrostomy or pancreatojejunostomy after pancreatoduodenectomy. Am J Surg 189(6): 720-729 [crossref]
31. Peng SY, Mou YP, Liu YB, Su Y, Peng CH, et al. (2003) Binding pancreaticojejunostomy: 150 consecutive cases without leakage. J Gastrointest Surg 7(7): 898-900 [crossref]
32. Peng SY, Wang JW, Lau WY, Cai XJ, Mou YP, et al. (2007) Conventional versus binding pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial. Ann Surg 245(5): 692-698 [crossref]
33. Chen XP, Huang ZY, Lau JW, Zhang BX, Zhang ZW, et al. (2014) Chen’s U-suture technique for end-to-end invaginated pancreaticojejunostomy following pancreaticoduodenectomy. Ann Surg Oncol 21(13): 4336-4341 [crossref]
34. Kennedy EP, Yeo CJ. (2011) Dunking pancreaticojejunostomy versus duct-to-mucosa anastomosis. J Hepatobiliary Pancreat Sci 18(6): 769-774 [crossref]