Perioperative Health Interventions in Children With Chronic Neuromuscular Conditions Undergoing Major Musculoskeletal Surgery: A Scoping Review

We performed a scoping review of articles published from January 1, 2000 through March 1, 2022 in which the authors evaluated the impact of a POHI on health outcomes for CCNMC undergoing major musculoskeletal surgery. Studies published before the year 2000 were excluded because we felt they may reflect aspects of perioperative care that are not relevant to current care. (PROSPERO CRD42022341653)

The main intervention, POHI, was any health care encounter or clinical intervention performed perioperatively with the aim of improving perioperative outcomes. The encounter or intervention had to include at least 1 clinical element of preoperative care. Examples of POHI include preoperative comprehensive health assessments; focused interventions on specific risk factors (eg, malnutrition, respiratory insufficiency, skin pathogens); and preoperative health service exposures to complex care, hospital medicine, and multidisciplinary providers.

All potential perioperative outcomes relating to the child’s preoperative health or postoperative recovery, any process measures, and perioperative utilization were considered.

A biomedical librarian searched 5 databases (PubMed, Embase, Cumulative Index of Nursing and Allied Health Literature, Web of Science, and the Cochrane Library) for articles published from January 1, 2000 through March 1, 2022. There were no language restrictions. All search strategies used controlled vocabulary terms from each database combined logically with relevant natural language keywords. Controlled vocabulary terms were chosen from within the subjects of neurologic and neuromuscular disease, health assessments, and the perioperative setting. Keywords consisted not only of general terms, such as “neuromuscular disease,” but also specific terms related to broader concepts, such as “cerebral palsy” and “muscular dystrophy,” along with variations of each disease. The same strategy was implemented for “health assessment” and elements of health assessments (eg, “functional status” and “surgical clearance”). All keywords were searched in specific title and abstract fields to maintain specificity. Gray literature searches were conducted in Google Scholar and ClinicalTrials.gov (Supplemental Information and Supplemental Table 4).

Using the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) criteria as a supporting framework²¹  and PRISMA Extension for Scoping Reviews: Checklist and Explanation,²²  2 study team members independently reviewed each title and abstract of the articles produced by the above search strategy to assess whether the study included outcome evaluation with a POHI exposure in CCNMC undergoing major musculoskeletal surgery. Two votes were required to exclude the study from further consideration. A third reviewer arbitrated discordant perceptions. The retained articles’ full texts were then reviewed by at least 2 team members to determine their appropriateness for study inclusion (Fig 1). Standardized data abstraction from the selected full-text articles was completed. The full-text articles’ bibliographies were then manually searched for additional articles. A total of 18 articles ultimately met the inclusion criteria for the present scoping review.

The risk of bias for each of the 18 retained articles was assessed by using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) assessment tool,²³  which is used to evaluate the risk of bias in 7 distinct categories: (1) bias relating to baseline confounding only, (2) bias in selection of participants into the study, (3) bias in classifications of interventions, (4) bias due to deviations from intended interventions, (5) bias due to missing data, (6) bias in measurement of outcomes, and (7) bias in selection of the reported result. Teams of 2 to 4 reviewers assessed each article’s risk of bias in each of these categories.

Because of the relatively small number of articles included in this scoping review, as well as the heterogeneity of the interventions and outcome measures, meta-analyses with pooled data were not performed.

Of the 7013 unique abstracts initially identified, 727 (10.4%) were retained for full-text review. Of these, 18 (2.5%) were retained for presentation in this review. Reasons for exclusion at the full-text review stage included no POHI (including at least 1 element of preoperative care) described (n = 602, 84.9%), neuromuscular population was not studied or stratified (n = 21, 3.0%), no health outcome was reported beyond orthopedic instrumentation or bone angle correction (n = 17, 2.4%), procedure did not include the manipulation of bones, muscles, or soft tissue (n = 2, <0.3%), and others (including those articles containing only abstracts; n = 67, 9.4%; Fig 1). The results of the 18 studies that were retained are presented below on the basis of the themes of their interventions and outcomes.

Four of the 18 retained articles assessed the impact of bundled perioperative interventions on SSI after spinal fusion for NMS in children. Three^12–14  used a pre- versus postintervention study design, and 1¹⁵  used a cohort comparison. Two were retrospective,^13,14  and 2 were prospective.^12,15  All 4 were single-center studies, with study sample sizes ranging from 160¹⁵  to 565.¹² 

Three studies included bundled pre- and intraoperative interventions, including preoperative skin decontamination and intraoperative antibiotic administration.^12,14,15  Glotzbecker et al and Salsgiver et al included the management of postoperative wound dressings. Glotzbecker et al and Tipper et al included perioperative antibiotic regimens, intraoperative blood salvage techniques, intraoperative patient body temperature control, and limits to surgical time. Glotzbecker et al included preoperative bowel preparation and a preoperative multidisciplinary conference case discussion. McIntosh et al included preoperative evaluation by a variety of clinicians (eg, pulmonology, respiratory therapy, anesthesiology), as well as baseline nutritional laboratories and pulmonary function tests.

The authors of all 4 studies reported a statistically significant reduction in SSI in patients exposed to the bundled interventions. Glotzbecker et al reported a deep SSI decrease from 8.3% to 0.9% (P = .005). Salsgiver et al reported that the underdosing of preoperative cefazolin or tobramycin was associated with an increased rate of SSI (47.1% vs 12.0%, P < .01), as was the underdosing of intraoperative cefazolin or tobramycin (40.0% vs 10.6%, P < .01). Tipper et al reported an SSI decrease from 8.6% to 2.2% (P = .0024). McIntosh et al reported 6.9% versus 0.0% SSI in children unexposed versus exposed to their intervention, respectively (P = .015; Table 1).

The authors of 6 articles evaluated perioperative interventions focused on optimizing specific organ function before surgery. Meltzer-Bruhn et al²⁴  and Obana et al²⁵  focused on preoperative nutrition; Takaso et al, across 3 studies that were closely related with respect to study population and content,^26–28  focused on preoperative respiratory function; and Vitale et al¹¹  focused on hematologic optimization. Three studies^11,24,25  were retrospective and used a control group, and 3^26–28  were prospective without a control group. All studies were single-center, with study sample sizes ranging from 14²⁷  to 155.²⁵ 

Obana et al investigated the impact of a nutritionist’s evaluation weeks before surgery. The evaluation included a review of anthropomorphic measurements, diet, and dietary recommendations. Meltzer-Bruhn et al investigated a nutrition evaluation by a gastroenterologist, registered dietician, or pediatrician within 1 year of surgery. Obana observed no postoperative differences in BMI between those who did versus those who did not have a nutritionist’s evaluation (multiple P values, all ≥.1), though they did note an increase in postoperative complications among those who had received the nutritionist’s evaluation (33% vs 16%, P = .03). Meltzer-Bruhn et al observed no difference in weight gain for children with versus without the nutrition evaluation (P = .9).

Takaso et al evaluated the effects of inspiratory muscle training via pulmonary function tests 3 and 6 weeks preoperatively in children with Duchenne muscular dystrophy (DMD). The authors observed that no patients required a prolonged ICU stay or ventilatory support and that forced vital capacity (FVC) increased from a mean of 22% to 26% (no P value reported) after inspiratory muscle training. Subsequently, at 2-year follow-up, no patients had sustained their increased FVC.

Vitale et al investigated recombinant human erythropoietin (rhEPO) administered subcutaneously 21, 14, and 7 days before surgery, as well as on the day of surgery. In addition, all patients received iron supplementation for 6 weeks before surgery. The authors noted that although both preoperative and discharge hematocrit levels were higher in patients receiving rhEPO and iron, there were no differences in transfusions, complications, or ICU length of stay (LOS; Table 2).

The authors of 8 studies focused on multidisciplinary approaches with various outcomes. Because of the heterogeneity of these studies, each is described individually below (Table 3).

Simon et al⁹  performed a retrospective analysis of orthopedic surgeons’ records describing the pre- and perioperative evaluation and management of children with medical complexity (CMC) undergoing spinal fusion, most of whom had significant neuromuscular disease. Preoperative pulmonology, cardiology, gastroenterology, and neurology consults were coordinated, and discharge planning meetings were organized. Pediatric hospitalists postoperatively comanaged 14 patients, of which 13 had NMS. Hospitalists implemented medication modification, nutritional adjustment, bowel regimen optimization, and care coordination. Mean LOS decreased from 8.6 days (95% confidence interval 8.0–9.2) to 6.25 days (95% confidence interval 5.5–6.9; P < .002) for all NMS patients, including those exposed and not exposed to hospitalist co-management.

Miller et al²⁹  performed a single-center, retrospective, preintervention (year 1999) versus postintervention (year 2008) analysis of CCNMC undergoing spinal surgery, wherein 31.5% of patients (23 of 73) followed a comprehensive preoperative algorithm that included nutrition, cardiology, and respiratory evaluations. These assessments occurred 3 to 6 weeks before surgery and were performed by a multidisciplinary team that included a nurse coordinator, respiratory therapist, physical therapist, social worker, case manager, dietician, and anesthesiologist. Perioperative planning included a discussion of discharge needs. The study authors found that the mean postoperative LOS was 19.0 versus 8.4 days in the pre- versus postintervention groups, respectively (P < .001).

Rappaport et al⁸  described the implementation of a hospitalist co-management program. The authors performed a single-center retrospective analysis comparing outcomes for 87 patients before the program was implemented, 40 patients while the program was partially implemented, and 80 patients after the program was fully implemented. Most patients had chronic medical problems or spastic quadriplegia, were technology-dependent, and took multiple medications at baseline. For pre-implementation patients, preoperative care was provided by a nurse practitioner or anesthesiologist, and postoperative care was provided by the ICU or orthopedic surgery team. For patients in the partially implemented program, post-ICU care included some involvement by a nurse practitioner or hospitalist. In the fully implemented program, pediatric hospitalists coordinated with the presurgical center and presented selected cases during a multidisciplinary conference. Preoperative interventions were not specified. The program was associated with increased BMI (P = .05), decreased anesthesia (P = .01) and surgery time (P = .001), and fewer days of postoperative parenteral nutrition required (P <.0006).

In a single-center retrospective case series, Rappaport et al³⁰  assessed how frequently pediatric hospitalists in the spinal surgery preoperative clinic made recommendations and evaluated the association between these recommendations and patient characteristics. At least 1 recommendation was made for 72.4% (155 of 214) of children, whereas the remaining 27.6% (59 of 214) received no recommendations before surgery. Children who were non-ambulatory, had seizures or gastrointestinal conditions, or took multiple chronic medications received more preoperative recommendations from the hospitalists.

Berry et al,³¹  in a retrospective analysis of health care claims for 1249 CMC undergoing spinal fusion, assessed the association of preoperative primary care visits with postoperative LOS. Surgery hospitalization cost was 12% lower (P = .05), and LOS was 9% shorter (P = .1) for children with 1 to 2 versus no preoperative visits. Surgery hospitalization cost was 21% lower (P = <.001), and LOS was 14% shorter (P = .01) for children with ≥3 versus no preoperative visits. Details of the visits’ contents were not reported.

In a single-center, retrospective, case-control analysis of CMC with NMS, Berry et al³²  examined preoperative visits with a pediatric complex care clinician. Approximately 49% (39 of 79) of children had a one-time, in-person health care encounter occurring weeks to months before surgery with a physician or advanced practice nurse affiliated with a complex care program. The visits involved a review of nutrition, coexisting conditions, chronic medications, and durable medical equipment and supplies, the identification of risk factors for surgery, and the development of preoperative care plans. The authors found that the rate of last-minute care coordination activities (ie, occurring in the days ahead of surgery) required for preoperative clearance was lower for children who had received a complex care assessment (1.8 vs 3.6 activities per patient, P < .001). The last-minute development of new perioperative care plans was also lower for these children (26% vs 50%, P < .002).

Deveza et al³³  compared complications among patients undergoing spinal instrumentation before versus after implementing a multidisciplinary care pathway. In this pathway, a multidisciplinary committee (composed of surgeons, pulmonologists, anesthesiologists, intensivists, hematologists, neuromonitoring experts, and dieticians) reviewed the indications for surgery, obtained pulmonary, nutrition, and anesthesia consultations, and completed a final preoperative assessment. Mortality, postoperative neurologic deficits, and SSI rates were compared between 107 patients having surgery before multidisciplinary pathway implementation and 109 patients having surgery after pathway implementation. Patients who experienced the multidisciplinary pathway had a higher preoperative BMI (P < .001), fewer postoperative neurologic deficits (P = .17), and decreased mortality 1 year post-operation (P = .04).

Sedra et al³⁴  reviewed the literature on the preoperative optimization of patients with NMS and recommended a multidisciplinary approach to address key preoperative concerns while formulating patient-specific treatment plans. Although this study met our inclusion criteria for review, the only published outcome included 1 patient’s perioperative experience.

With this scoping review, we identified 18 studies that assessed the impact of POHIs on outcomes for CCNMC undergoing major musculoskeletal surgery. Three major themes emerged from these studies: (1) using perioperative intervention bundles to prevent SSI, (2) using focused preoperative interventions to optimize specific organ function, and (3) using preoperative visits and multidisciplinary interventions incorporating hospitalists, complex care clinicians, and others to optimize perioperative health and safety. In our own perioperative experiences, we struggle with which health aspects to prioritize for our CCNMC undergoing major musculoskeletal surgery. The 18 studies included in this scoping review reflect institutions’ real-world attempts to optimize the perioperative outcomes of CCNMC. It is encouraging that the authors of many articles reported a positive impact of POHI on outcomes, including fewer postoperative complications.

The authors of all 4 studies within this theme used various bundled interventions (eg, preoperative skin decontamination, intraoperative antibiotics, and patient temperature control) to decrease SSI.^12–15  Children’s Hospitals’ Solutions for Patient Safety reports that SSI represent the fourth largest harm for hospitalized children.³⁵  As such, there is an ongoing need to increase innovative interventions (eg, decreased traffic during surgery, nasal decolonization, negative pressure wound therapy) to decrease SSI³⁶  in CCNMC.

The studies examining the effects of preoperative optimization of specific organ function revealed that the interventions had no significant effect on outcomes. The nutritional studies of Meltzer-Bruhn et al²⁴  and Obana et al²⁵  did not reveal any significant difference in preoperative BMI, although it is important to note that BMI may be a flawed proxy for assessing nutritional status^37,38  in CCNMC. Vitale et al’s¹¹  optimization of hematologic function did not change important surgical outcomes. Takaso et al^26–28  identified a small cohort of DMD patients treated preoperatively with pulmonary resistance training who were able to undergo surgeries with minimal complications, although these FVCs gains were not sustained 2 years after surgery. Despite these seemingly negative results, these studies nevertheless offer insights that can direct further research.

The preoperative and multidisciplinary assessments within this review were often conducted weeks to months before surgery, presumably to ensure that enough time remained to enact recommendations and achieve the desired benefits. Many studies conveyed specific details of their POHI,^{9,15,29,32,33}  although 1³¹  shared no intervention details. Many assessments evaluated nutrition, coexisting health conditions, airway clearance regimens, respiratory and hematologic functions, and technology dependence.^{8,9,15,29,32,33}  Patients with special health care needs have increased odds of unmet health care goals,^39,40  supporting the need for a primary medical or surgical home that can (1) provide routine and health care maintenance,^41,42  (2) proactively monitor and comprehensively optimize perioperative health,⁴³  and (3) support families in their transition home after surgery.⁴⁴  Expert clinicians that provide POHI may result in fewer SSI; decreased last-minute care coordination needs; and reduced postoperative LOS, mortality, and spending. The 18 studies examined in this review involved different provider types performing POHI. Some used multidisciplinary teams to perform the assessment,^{8,9,14,15,29,32–34}  1³²  used a single complex care provider, 3 relied on a respiratory therapist,^26–28  and 1 did not specify who provided the preoperative interventions.³¹ 

A risk of bias assessment involves comparing a study to a theoretical optimally performed study, typically a randomized control trial. A total of 11 of the 18 studies included in this review were deemed to have either a serious or critical risk of bias (Supplemental Tables 5–8). Most studies included small numbers of patients and did not reveal sample size calculations or statistical power. Rather than being evidence of poor study design, this underscores the profound challenges involved in effecting meaningful change in the care of these patients. Moreover, it highlights the ongoing need to investigate opportunities to improve the multifaceted care of CMC, a vulnerable and growing population.

Given the breadth of care that CCNMC receive, no single approach to POHI is likely to be effective for all children. Instead, each institution should consider how they can best optimize patients for the entirety of their perioperative experience rather than simply ensuring preoperative surgical clearance. No studies in the present review revealed descriptions or data regarding the shared decision-making processes with patients and caregivers about whether and when to proceed with surgery despite significant risks. Shared decision-making will be important to highlight in future studies because parents of CCNMC often desire better preparation for recovery.⁴⁵ 

The small number of studies meeting our inclusion criteria and the heterogeneity in their interventions and outcomes precluded our ability to perform a pooled analysis. In addition, the disparate study designs do not support clear and specific evidence-based recommendations about POHI targeting CCNMC, except that some bundled collaborative interventions may yield better outcomes. This review does not address issues of health equity, although ensuring equitable access to interventions intended to optimize perioperative health and safety is paramount. Most studies did not reveal the cost of their interventions, so we are largely unable to assess their financial impacts. Certainly, there are opportunities for authors of future studies to investigate these topics further.

Despite these limitations, the findings from this review reveal the promise of POHI in improving perioperative care for CCNMC vis-à-vis care coordination, LOS, hospitalization costs, and SSI. Multicenter prospective studies are needed to better understand the impact of specific POHI in improving the quality of care in CCNMC undergoing major musculoskeletal surgery.