Effects of Comprehensive Geriatric Care Models on Postoperative Outcomes in Geriatric Surgical Patients

A Systematic Review and Meta-analysis

Aparna Saripella; Sara Wasef; Mahesh Nagappa; Sheila Riazi; Marina Englesakis; Jean Wong; Frances Chung

Disclosures

BMC Anesthesiol. 2021;21(127) 

In This Article

Discussion

To our knowledge, there was no previous systematic review or meta-analysis on comprehensive geriatric care models and their effects on postoperative outcomes in elderly population undergoing surgery. Our systematic review and meta-analysis yielded eleven studies on geriatric care models with CGA as the major component. We found these care models did not show a significant difference in the prevalence of delirium, LOS, 30-days readmission rates, and 30-day mortality.

Characteristics of Comprehensive Geriatric Care Model

Of the eleven studies[10,11,13,15–19,22–24] most adapted six comprehensive geriatric care models.[11–16,18,19,33] This systematic review allowed for a comparison of the similarities and differences in the different care models. All geriatric care models used CGA. CGA identifies the co-morbid conditions of elderly patients, thus helping health care professionals optimize perioperative care. It allows the opportunity for counselling patients regarding risk reduction and evaluation of nonsurgical treatment options.[34] Most of the programs centred around management of symptoms of cognitive impairment, frailty, and immobility, showcasing the importance of managing impairment before surgery as a preventative measure and after surgery to improve postoperative adverse outcomes.

ERAS pathways outline preoperative and postoperative management, supplemented by intraoperative risk reduction efforts such as specific administration of analgesics, nausea and vomiting, and hemodynamic management,[35,36] The American Society for Enhanced Recovery and Perioperative Quality Initiative and the European Society of Anaesthesia have echoed the importance of assessing preoperative delirium risk and informing patients of their status before surgery.[34,35] Thus, both ERAS and geriatric care models use delirium assessment and management. Our review excluded ERAS pathways because they differ from geriatric care models which put a primary emphasis on CGA. Also, the involvement of the geriatric team for CGA is important besides a multidisciplinary team of surgeons, anaesthesiologists, pharmacists, occupational therapists/physiotherapists, and nutritionists. The multidisciplinary team approach has been credited for the disciplined implementation of the care model.[18]

Outcome

Although the prevalence of delirium was 2.1% less in the intervention group vs the control group in six studies, it was not significant statistically. However, in the three RCTs, the prevalence of delirium was reduced by over 50% in the intervention group, with an absolute risk reduction of 8% and Number Needed to Treat (NNT) of 13. The high prevalence of delirium and its correlation with diminished quality of life and cognitive impairment emphasizes the importance of selecting it as a clinical outcome. Both the Hospital Elder Life Program (HELP),[14] and Proactive care of Older People (POPS),[17] care models found significant reduction in the prevalence of delirium. Both referred patients to specialists following CGA and offered a comprehensive care model towards improvement of delirium.[10,17] On the contrary, the other models provided only recommendations for recovery after assessment.[13,15,19,22] This may have accounted for the differences in the prevalence of delirium in these care models.

The LOS is a common clinical outcome to evaluate the effectiveness of interventions.[23] A shortened LOS leads to decreased cost and reduction in the probability of contracting infections.[37,38] Although variation in measurement of outcomes limited the ability of pooled estimate among the RCTs, all the programs showed a reduction in LOS except of one using the Liaison Intervention in Frail Elderly (LIFE) model.[13]

Readmission to hospital has gained popularity as an outcome measure in quality improvement. Literature is divided over its validity due to the inconsistent correlation between hospital readmission and other clinical outcomes.[39] Our meta-analysis results showed no significant differences in 30-day readmission between the intervention and control groups. The only geriatric care model showed significant result was the Perioperative Optimization of Senior Health (POSH) model.[15] Although our review showed that CGA only decreases the prevalence of delirium by over 50% in elderly patients, Eamer et al.[40] found CGA decreased mortality in hip fracture patients, and Ellis et al.[26] showed the CGA improved patients being able to live at home.

We rated the quality of the evidence for outcomes using GRADE system.[41] The ratings were very low for all the outcomes when we combined RCTs and non-RCTs, (indicating considerable uncertainty regarding the estimates of effect). In our research, we found that there was no significant difference among intervention and control groups. However, further RCTs of adequate power and clearly defined endpoints in specific surgical procedures are warranted to determine the overall benefits in the postoperative outcomes.

Limitations

There are some limitations in this systematic review and meta-analysis. There are only four RCTs with 7 non-RCTs, and the studies were heterogeneous with different types of surgeries. Eight studies were single centre studies[10,11,15–17,19,22,24] with three multi-centre studies.[13,18,23] The chance of bias increases as the staff between the intervention and control groups is the same augmenting the chance of contamination. Also, including pre-post design interventions introduces bias due to the lack of randomization. Furthermore, the multidisciplinary approach varied from between studies as they were conducted in different institutions and at different time periods. This may cause lack of uniformity in the intervention and may have contributed to another source of bias. The studies included were in English and we may have missed studies in other languages. Nevertheless, this systematic review provides a summary of existing evidence and serves as an impetus for the academic community to do further research in this area.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....