Introduction

Laparoscopic bariatric surgery is an effective therapy for morbid or complicated obesity, and the fast-track approach is becoming popular [1, 2]. However, there are serious perioperative challenges and factors that may hamper the perioperative outcome or recovery of bariatric surgery patients [3, 4]. Postoperative pain may be significant and has negative impact on recovery, respiration, haemodynamics, mentation, ambulation, bowel function and length of hospital stay. Also, opioid analgesia in obese patients is associated with serious adverse effects such as hypopnoea, hypoxaemia, drowsiness, ileus, vomiting, delayed ambulation and mortality [5, 6]. These complications are compounded by pre-existing obstructive sleep apnoea (OSA), cardiorespiratory insufficiency, psychiatric disorder and chronic opioid use which may be present in obese patients [7–9]. Thus, optimal perioperative, non-opioid, multimodal analgesia may expedite fast-track bariatric surgery. Few small retrospective studies of postoperative analgesia in bariatric surgery have been published [10–13]. However, there are no studies of the broader topic of perioperative analgesia and the overall impact. This large prospective study examined the overall impact of perioperative analgesia on fast-track bariatric surgical care.

Methods

This prospective observational study was approved and registered by the research department of Central Manchester University Hospital, Manchester, UK. Perioperative patient outcome data were recorded from January 2007 to January 2013. Consecutive patients underwent laparoscopic Roux-en-Y gastric bypass (LRYGB) under general anaesthesia. LRYGB was performed by two experienced surgeons, who are contemporaries and use similar surgical techniques. Surgery involved five trocar sites (three 5-mm and two 12-mm), hand-sewn gastrojejunal anastomosis and suture of all trocar sites. No concurrent or other surgery was performed. Operative time was similar for the patients, with mean duration of 100 min and median duration of 95 min.

Balanced general anaesthesia, including prophylactic anti-emetic therapy, was provided by three experienced anaesthesiologists. Supplemental intraoperative analgesia was administered based on the clinical decision and preference of each anaesthesiologist. All patients received surgical site infiltration with bupivacaine + epinephrine at the start and end of surgery. In the postanaesthesia care unit (PACU), analgesia consisted of intravenous (IV) morphine patient-controlled analgesia (PCA). On the surgical ward, analgesia comprised IV acetaminophen 1 g 6-hourly regularly and morphine PCA.

General data collected included patients’ gender, age, body mass index (BMI), comorbidities, obstructive sleep apnoea (OSA) diagnosis, American Society of Anesthesiologist (ASA) physical status, perioperative monitoring, surgery duration, perioperative complications, PACU duration, intensive care unit duration and length of hospital stay. Specific perioperative data collected included intraoperative analgesic, numeric pain score at PACU and 24 h, opioid analgesia requirement at PACU and 24 h, postoperative nausea/vomiting (PONV) and time to sustain oral drinks. Opioid requirement was measured by the dose, demand and duration of PCA used. Pain was categorized based on numeric pain score: 0 = no pain, 1–3 = mild pain, 4–6 = moderate pain and 7–10 = severe pain. Obesity was categorized based on BMI: severe obesity = 36–39.9, morbid obesity = 40–49.9 and super obesity is ≥50.

Data were analysed using SPSS® version 20 (IBM Corp., Armonk, NY, USA) and included descriptive statistics. Bivariate analysis was performed using the Student’s t test. Differences between groups were compared using the Pearson’s chi-square test or the Fisher’s exact test as appropriate. p value <0.05 was considered statistically significant.

A total of 412 consecutive patients were studied over a 6-year period. The BMI and gender distribution is shown in Table 1 with BMI range of 38–80, mean of 49 (±7), median of 48 and male:female ratio of 1:4.

Table 1 BMI and gender distribution; count (percentage of total)
Full size table

All the patients were contemporaneous adults from a homogeneous population. The age range was 18–64 years, mean age was 42 (±9) years and median age was 42 years. The age distribution is shown in Table 2.

Table 2 Age distribution
Full size table

Results

Approximately 82% of patients received multimodal intraoperative analgesia, comprising IV opioid, acetaminophen, with or without non-steroidal anti-inflammatory drug (NSAID). The opioid administered in 83% of patients was morphine and tramadol in 17%. About 48.5% received morphine 0.1 mg/kg up to maximum of 20 mg + acetaminophen 2 g + parecoxib 40 mg; 17.2% received morphine 0.1 mg/kg + acetaminophen 2 g; 17% received tramadol 1.5 mg/kg up to maximum of 200 mg + acetaminophen 2 g + diclofenac 75 mg; and 17.2% received morphine 0.1 mg/kg (Table 3).

Table 3 Intraoperative IV analgesia
Full size table

Postoperative Pain & Duration in PACU

Pain assessment in PACU showed that 4% had severe pain, 7% had moderate pain and 63% had mild pain (Table 4). However, 26% had no pain; this group comprised all patients who received tramadol + acetaminophen + diclofenac (65% of pain-free patients) and some patients who received morphine + acetaminophen + parecoxib (35% of pain-free patients). In the PACU, 26% of patients did not use any analgesia; this group comprised all the patients who received tramadol + acetaminophen + diclofenac (65%) and some patients who received morphine + acetaminophen + parecoxib (35%). Multimodal intraoperative analgesia provided lower PACU pain scores compared to unimodal intraoperative morphine analgesia (p = 0.0001). Multimodal analgesia comprising tramadol + acetaminophen + diclofenac provided lower PACU pain scores than other multimodal analgesia regimens (p = 0.0001).

Table 4 Pain scores in PACU vs intraoperative analgesia type
Full size table

PACU duration was categorized as short ≤50 min, medium = 51-100 min, long = 101–150 min and longest ≥ 151 min. PACU duration varied with intraoperative analgesia type, as shown in Table 5. Approximately 67% of patients spent short duration in PACU; this group mainly comprised patients who received multimodal analgesia, and they had shorter PACU duration than patients who received unimodal morphine analgesia (p = 0.01). Multimodal intraoperative analgesia comprising tramadol + acetaminophen + diclofenac was associated with shorter PACU duration compared to other analgesia types (p = 0.001).

Table 5 PACU Duration vs intraoperative analgesia type
Full size table

Postoperative Pain and Analgesia Requirement at 24 h

Patients had lower pain scores at 24 h compared to PACU (p = 0.001). Pain assessment at 24 h showed that 1.5% had severe pain, 5.6% had moderate pain, 68.9% had mild pain and 24% had no pain (Table 6). The pain-free group comprised all the patients who received tramadol + acetaminophen + diclofenac (74% of pain-free patients) and some patients who received morphine + acetaminophen + parecoxib (26% of pain-free patients). Patients who received multimodal intraoperative analgesia had lower pain scores at 24 h compared to patients who received unimodal morphine analgesia (p = 0.002). Analgesia usage at 24 h showed that 24% did not use the PCA which was provided, 73% discontinued PCA on the first postoperative day and 3% discontinued PCA on the second day. Patients who did not use PCA comprised those who received tramadol + acetaminophen + diclofenac (74%) and some patients who received morphine + acetaminophen + parecoxib (26%). Patients who received multimodal intraoperative triple analgesia had lower pain scores at 24 h and shorter use of PCA compared to other patients (0.002).

Table 6 Pain scores at 24 h vs intraoperative analgesia type
Full size table

Postoperative Oral Intake and PONV

PONV occurred in 19.2% of patients and varied with intraoperative analgesia type (Table 7). Patients who received unimodal intraoperative morphine analgesia comprised 33% of PONV cases. Multimodal intraoperative analgesia was associated with lower PONV rates compared to unimodal morphine analgesia (p = 0.001).

Table 7 Postoperative nausea/vomiting (PONV) at 24 h vs intraoperative analgesia type
Full size table

About 90% of patients had sustained oral drinks postoperatively on the day of surgery (day 0), and this varied with intraoperative analgesia type (Table 8). Patients who received multimodal intraoperative triple analgesia had earlier and better sustained postoperative oral drinks than other patients (0.004).

Table 8 Time to sustain oral drinks vs intraoperative analgesia type
Full size table

Postoperative Respiratory Complications

About 1.7% of patients had early postoperative hypopnoea which required respiratory support with mask-CPAP (continuous positive airway pressure); and 0.4% of patients required supplemental nasal oxygen after 24 h (Table 9). There was no significant association between postoperative respiratory dysfunction and intraoperative analgesia type (p = 0.9). Half of the patients with respiratory dysfunction were successfully treated with single dose of IV doxapram 75–100 mg, which enabled prompt recovery and discharge. Compared to patients who received intraoperative tramadol, patients who received intraoperative morphine had significantly higher rates of postoperative doxapram therapy (p = 0.04).

Table 9 Postoperative respiratory dysfunction vs Intraoperative analgesia type
Full size table

Length of Hospital Stay

Majority of patients (59%) were discharged home at 24 h. Length of hospital stay was associated with intraoperative analgesia type (Table 10). Multimodal intraoperative analgesia comprising tramadol + acetaminophen + diclofenac was associated with earlier ambulation and shorter hospital stay than other analgesia types (p = 0.01).

Table 10 Length of hospital stay vs intraoperative analgesia type
Full size table

Discussion

Postoperative pain and analgesia are associated with challenges in bariatric surgery patients [3, 6, 8, 9]. Multimodal analgesia has better efficacy and less complications [11–14]. About 82% of patients in our study received multimodal intraoperative analgesia, which enhanced their overall good perioperative outcome. It was associated with better postoperative analgesia, shorter PACU duration and lower postoperative opioid requirement. This confirms that multimodal analgesia is beneficial and essential in bariatric surgery patients and corroborates other studies [11–14]. Our study demonstrated that multimodal analgesia is associated with less PONV, earlier postoperative oral intake, earlier ambulation and shorter hospital stay, and these are highly essential factors for fast-track or enhanced recovery bariatric surgical care. Our study also highlights the economic benefits of multimodal analgesia with shorter PACU duration, lower postoperative opioid requirement, earlier ambulation and shorter hospital stay.

Multimodal analgesia involves the simultaneous use of different analgesics that act via different physiologic processes to produce effective synergistic analgesia with less complications. In our study, it involved various combinations of intraoperative bupivacaine infiltration and IV acetaminophen, morphine, tramadol, parecoxib or diclofenac. Bupivacaine is a local anaesthetic which temporarily blocks neurotransmission, thereby providing analgesia for 4–8 h. Every patient in our study received bupivacaine infiltration, which contributed to multimodal analgesia, and this corroborates other studies [10, 15, 16].

Morphine is an efficacious opioid analgesic but is associated with significant adverse effects including airway or respiratory compromise, which may be severe in bariatric patients [3, 6, 9, 11, 12]. About 83% of our patients received intraoperative morphine, which provided good analgesia especially as part of multimodal analgesia, but was associated with postoperative hypopnoea. Our study confirms that morphine-related hypopnoea may be significant in bariatric patients but can be treated with the analeptic doxapram, and this corroborates a previous study [9]. Postoperative morphine was administered in the form of PCA, which afforded smaller doses and less complications, and contributed to lower pain scores at 24 h.

Tramadol is a synthetic analgesic with opioid, noradrenergic and serotoninergic effects. Its opioid and non-opioid analgesic effects are synergistic, thereby producing effective analgesia and minimal serious side-effects [17]. About 17% of our patients received intraoperative tramadol, and they had better postoperative analgesia, shorter PACU duration, less postoperative opioid requirement, earlier ambulation, shorter hospital stay, less postoperative hypopnoea and less doxapram requirement compared to patients who received morphine. These are very positive and essential perioperative outcome measures in bariatric patients. Our study is the first to highlight that tramadol analgesia is most suitable, efficacious, safe and associated with the best perioperative outcomes in bariatric patients. The main side-effect of tramadol is vomiting, but our study showed that this can be effectively prevented by pre-emptive multimodal anti-emetic therapy [17].

Acetaminophen is an analgesic with opioid-sparing activity but has rare adverse effects [11–13, 18]. It is an essential component of multimodal perioperative analgesia and was used in every case of multimodal analgesia in our study. Our study confirms that multimodal analgesia involving acetaminophen, reduces opioid requirement and side-effects but without acetaminophen-related adverse effects as indicated in other studies [11–13, 18].

Diclofenac and parecoxib are non-steroidal anti-inflammatory drugs (NSAID) and opioid-sparing analgesics with few adverse effects [12, 16, 19–21]. Two thirds of patients in our study received NSAID, which enhanced multimodal analgesia. Our study confirms that multimodal analgesia involving NSAID, reduces opioid requirement and side-effects but without NSAID-related adverse effect as shown in other studies [12, 16, 19–21].

Our study is a reliable prospective observational clinical research with good sample size and valid results. However, it is limited by the time-consuming consecutive sample of patients. A larger sample size may probably reveal more perioperative complications. Further larger studies would be interesting, and should focus on pain-related and analgesia-related perioperative complications in bariatric surgery patients.

Conclusion

Postoperative pain and analgesia pose peculiar challenges in bariatric surgery patients. However, multimodal intraoperative analgesia is a good foundation towards optimal postoperative analgesia. Multimodal intraoperative analgesia provides better postoperative analgesia, less complications and better perioperative outcomes and facilitates fast-track perioperative care. Tramadol is suitable, efficacious, safe and associated with the best perioperative outcomes in bariatric surgery patients.