Retrospective analysis of 30-day unplanned readmission after major abdominal surgery with reversal by sugammadex or neostigmine

Oh TK, Oh AY, Ryu JH et al.

British Journal of Anaesthesia 2019; 122(3):370-378

Background

• The RECITE Study (2015), a Canadian prospective multicentre study, demonstrated the incidence of residual neuromuscular blockade (Train of Four ratio <0.9) to be 64% at the time of tracheal extubation and 57% at the time of arrival to recovery.
• Increasing evidence suggests the use of neostigmine/glycopyrrolate as a reversal agent may be associated with increased post-operative complications, either due to residual neuromuscular block, neostigmine-induced muscle weakness or cardiovascular side effects and complications, leading to increased hospital length of stay and unplanned hospital readmission.
• Sugammadex is associated with fewer postoperative complications, but its impact on 30-day unplanned readmission is unclear.
  • A reduction of readmissions is important for improving patient care and lowering costs; readmissions after surgery are closely associated with postoperative complications.
• Approximate costs of relevant treatments and procedures:
  • Emergency laparotomy: £13,000 per patient
  • Right hemicolectomy: £4,000 (laparoscopic) / £3,400 (open) per patient
  • Neostigmine/Glycopyrulate 2.5mg/0.5mg: £1.50 per ampoule
  • Sugammadex 200mg: £59.60 per vial

Aim of Study                                                                                                                      

• Determine the impact of sugammadex on 30-day unplanned readmission.

Design/Location

• Single-centre retrospective observational study of adult patients after elective major abdominal surgery between 2010 and 2017, where rocuronium was the only neuromuscular blocker used.
• Seoul National University Bundang Hospital, Seongnam, Republic of Korea.
  • A tertiary teaching hospital.

Primary Outcome

• Difference in incidence of 30-day unplanned readmission between reversal with sugammadex or neostigmine.

Secondary Outcomes

• Length of hospital stay after surgery.
• Related hospital charges (total charges excluding those related to surgery and anaesthesia).

Methodology/Statistics

• Analysis included propensity score matching and generalised mixed-effects modelling.
• Mixed-effects logistic regression analysis of 1479 patients (sugammadex: 355; neostigmine: 1124).

Results

In the sugammadex group, compared with the neostigmine group:

• Incidence of 30-day unplanned readmission was 34% lower (odds ratio [OR]: 0.66, 95% confidence interval [CI]: 0.46-0.96, P=0.031)
• The length of hospital stay was 20% shorter (exponential regression coefficient: 0.80, 95% CI: 0.77-0.83, P<0.001)
• Related hospital charges were 24% lower (exponential regression coefficient: 0.76, 95% CI: 0.67-0.87, P<0.001)

For patients living ≥50 km from the hospital:

• Incidence of 30-day unplanned readmission was 68% lower in the sugammadex group than in the neostigmine group (OR: 0.32, 95% CI: 0.13-0.79, P=0.014).
• It was not significant for patients living <50 km from the hospital (P=0.319).

Conclusions

• Compared with neostigmine, reversal of rocuronium with sugammadex after major abdominal surgery was associated with a lower incidence of 30-day unplanned readmission, a shorter hospital stay, and lower related hospital charges.

Discussion/Limitations from the Study

• Regarding the significance of the group living >50km from the hospital:
  • This study counted all 30-day unplanned readmissions after discharge.
  • Many patients could have been admitted to the institution for other medical illnesses, rather than postoperative complications.
  • Patients living far away that experienced illnesses other than surgical complications might be hospitalised in nearby hospitals whilst patients with surgical complications likely returned for treatment regardless of the distance.
  • Therefore, the effect of sugammadex on post-operative complications might be more evident among patients who lived far from the hospital.
• Retrospective observational study
  • May have resulted in selection bias and impaired data quality or accuracy compared with a prospective study.
• A single centre study looking at elective surgery only, involving mostly ASA 1 and 2 patients.
  • May have compromised the generalisability of the findings.
• The proportion of patients who underwent NMB reversal with sugammadex was relatively low at 10.7%.
  • May have affected the Power of the study, contributing to the non-significant result for patients living <50km from the hospital.
• Did not count cases of unplanned readmission to other hospitals.
  • Could have resulted in bias.
• Counted all 30-day unplanned readmissions after discharge.
  • Many patients could have been admitted for other medical illnesses rather than postoperative complications.

Strengths of the Study

(taken from the corresponding editorial, “Does reversal of neuromuscular block with sugammadex reduce readmission rate after surgery?”; Bose, Xu and Eikermann)

• Use of generalised mixed-effects regression analysis to adjust for the random effects of providers.
  • Ignoring the clustering feature of the data could lead to overestimation of the significance of the exposure effect.
  • By using a random effects model, the authors obtained the appropriate estimation and P values.
• Use of matched sets of patients through propensity score matching to make additional attempts to control for unadjusted confounding.
  • These otherwise might have affected the exposure (reversal agent) assignment.
  • Most of the patients who received sugammadex were matched and additional sensitivity analysis was performed using the entire unmatched cohort to further corroborate their findings.

Discussion from Journal Club Meeting

• The significant incidence of residual neuromuscular blockade at extubation and in recovery found in the RECITE study is a concern, as is the evidence of complications associated with the use of Neostigmine/Glycopyrulate, particularly in high risk patient groups.
• Sugammadex is associated with a better safety profile than neostigmine, particularly regarding residual neuromuscular block after reversal.
• It also reverses various degrees of residual neuromuscular more quickly.
• Sugammadex use may subsequently lead to a lower incidence of residual neuromuscular block and associated side-effects.
• Analysis of unplanned readmission with retrospective observational studies is complicated this study has its limitations as described above.
• More studies are needed to support its conclusions, particularly looking at higher risk patient groups and emergency situations.
• Inclusion of more patient-centred outcomes such as postoperative morbidity (e.g. pulmonary) and mortality would significantly strengthen the overall message of the paper and future studies.

 

Summary by Dr P Daum. Journal Club Meeting 25^th July 2019.