Effect of Osteopathic Manipulative Treatment on Incidence of Postoperative Ileus and Hospital Length of Stay in General Surgical Patients

Methods

After presenting our study as retrospective and noninterventional and the data set as encrypted, we received institutional review board exemption. We then retrospectively reviewed the electronic medical records of all general surgical patients who underwent major gastrointestinal operation (defined as small- or large-bowel resection or gastric resection or repair) at a 350-bed urban community hospital from January 1 through December 31, 2011. Demographic data; American Society of Anesthesiologists (ASA) physical status class¹³; preoperative comorbid conditions; types of operations performed; postoperative complications; postoperative time to flatus, clear liquid diet, and bowel movement; postoperative hospital LOS; postoperative electrolyte abnormalities; and types of narcotics used were tabulated from the electronic medical records. The ASA class, which classifies physical status on a scale of 1 to 6 (1, healthy; 6, brain dead), served as a standardized measure of perioperative complication risk. The only additional inclusion criterion was a complete data set at the time of chart review, and the only exclusion criterion was death.

We reviewed progress notes for the use of OMT and categorized patients into 2 groups: those who received OMT during the early postoperative course (OMT group) and those who did not receive any OMT (non-OMT group). We also noted the type of OMT techniques used (eg, muscle energy, myofascial release), the duration of the OMT session, the body areas treated, and whether the OMT was performed by physicians, residents, students, or a combination.

We compared tabulated data between the groups by using 1-way analysis of variance and used StatPlus software (Mac version 4.8.0; AnalystSoft Inc, Vancouver, British Columbia, Canada) to perform statistical analyses. We used a P value of .05 as a standard cutoff for statistical significance.

Results

Of 69 patients who underwent a major gastrointestinal operation during the review period, 61 (88%) had complete data sets at the time of chart review. Six (10%) of these patients had died and were thus excluded from the study. Data for the remaining 55 patients were categorized and compared. Seventeen of these patients (31%) received postoperative OMT in addition to patient care dictated by attending physicians; the remaining 38 patients did not receive postoperative OMT (Figure 1).

Figure 1.

Flow diagram of patients who received or did not receive osteopathic manipulative treatment (OMT) after gastrointestinal operation. Eligible patients were defined as those who underwent small- or large-bowel resection or gastric resection or repair. Patients who died were excluded from the study.

The mean (standard deviation [SD]) age of the patients was 60.3 (17.7) years in the OMT group and 62.1 (15.8) years in the non-OMT group (P=.70), and the mean (SD) ASA class in these groups was 2.5 (0.6) and 2.7 (0.7), respectively (P=.31) (Table 1). The numbers of preoperative comorbid conditions, postoperative complications, and electrolyte abnormalities were similar in the 2 groups. Morphine sulfate was the only parenteral narcotic used in both cohorts (Table 2).

The operations performed included small- and large-bowel resection and gastric resection or repair, and the percentages of patients who underwent each type of operation were similar in the 2 groups (Table 2).

The times to bowel movement and clear liquid diet did not differ significantly between the 2 groups. The mean (SD) time to flatus was 3.1 (0.6) days in the OMT group and 4.7 (0.4) days in the non-OMT group (P=.035). The postoperative hospital LOS was also significantly different between groups, with a mean (SD) LOS of 6.1 (1.7) days in the OMT group and 11.5 (1.0) days in the non-OMT group (P=.006) (Table 1 and Figure 2).

Figure 2.

Mean (standard deviation) postoperative days to flatus and postoperative hospital length of stay (LOS) for general surgical patients who did or did not receive postoperative osteopathic manipulative treatment (OMT).

Each patient who received OMT underwent a single treatment session performed by a surgical resident within 48 hours of the major gastrointestinal operation. The durations of treatment ranged from 15 to 35 minutes, and the types of treatment ranged in amount of force from cranial manipulation to direct myofascial release. Frequently treated areas included the costophrenic and costovertebral areas and the cervical spine.

Comment

Multiple review articles have suggested that a multimodal approach should be used to prevent and manage POI.^2,14 We compared 2 similar groups of general surgical patients and demonstrated that the use of OMT within 48 hours after a major gastrointestinal operation is associated with less postoperative time to flatus and a shorter postoperative hospital LOS.

The mechanism of action of OMT in this patient population is unclear. Osteopathic manipulative treatment has a known modulatory effect on the autonomic nervous system,⁹ and recent studies have shown that OMT may have distinct biochemical and anesthetic effects. Salamon et al¹⁵ suggested that OMT may function by increasing nitric oxide levels in the blood. Nitric oxide production has the potential to attenuate cellular damage from intestinal ischemia and may improve healing.¹⁶ McPartland et al¹⁷ suggested that OMT may elicit cannabimimetic effects, which in the gastrointestinal system include modulation of intestinal motility.¹⁸ Postoperative pain levels and the amount of narcotics used have also been shown to decrease after OMT.¹⁹ Decreased pain and opioid use may lead to earlier ambulation and improved intestinal motility. Taken together, the biochemical and neurologic effects of OMT may affect the gastrointestinal tract during the postoperative period and allow more rapid return of bowel function.

The present study demonstrated that patients receiving OMT had a shorter postoperative hospital LOS, an important component of hospital costs. Although it is complicated to quantify the absolute cost of prolonged LOS, a mean decrease of 5.4 inpatient days represents a substantial improvement in cost-effectiveness. Further studies directly analyzing the cost-effectiveness of OMT are warranted.

In accordance with the osteopathic medical curriculum for the surgical residency, our institution permits the use of OMT for any patient but has no specific OMT protocols. Similarly, all general surgery attending physicians (allopathic and osteopathic) encourage osteopathic residents to use OMT but do not require its use. Our review demonstrates that some general surgery residents perform postoperative OMT and that the use of a single OMT session is associated with improved outcomes. Similar to findings in what is, to our knowledge, the only other recent study of OMT and POI,¹² our study demonstrated an association between the use of OMT and a decrease in postoperative hospital LOS. In their study, Crow and Gorodinsky¹² analyzed findings in patients who had undergone any abdominal surgery, demonstrated a mean age difference between the cohorts (ie, OMT and non-OMT), and did not measure time to return of bowel function. Our study analyzed a more homogeneous population of patients who underwent only major gastrointestinal operations and in whom there was no statistically significant difference in mean age between treatment cohorts; it also demonstrated a statistically significant difference between cohorts in time to return of bowel function.

Our study has some important limitations. First, because of its retrospective design, we are unable to demonstrate whether OMT caused the differences observed between treatment cohorts. A prospective, randomized, blinded study, similar to that of Mills et al,²⁰ would be necessary to establish a cause-effect relationship. The addition of a “sham-OMT” group, such as a light-touch protocol group of a study, would also help distinguish placebo effect.

Second, our sample size was small because of patient death or lack of complete data sets, allowing analysis of findings in only 80% of all patients who underwent major gastrointestinal operations during the period studied. Larger studies demonstrating an earlier return of bowel function with OMT would help validate our results. Third, without standardized use of OMT, there is a potential for selection bias. However, patient age, ASA physical status class, comorbid conditions, and operation type were similar in the 2 cohorts, suggesting that the use of OMT was spread evenly across the study population.

Conclusion

Osteopathic manipulative treatment applied after major gastrointestinal operation is associated with decreased time to flatus and decreased postoperative hospital LOS. We recommend that general surgeons consider the use of postoperative OMT.