Background

Adhesive small bowel obstruction (ASBO) is one of the most common and significant complications after abdominal surgery,[1-5] with an incidence rate as high as 2.4%.[1] Adhesions are abnormal fibrous bands between organs and/or tissues that are normally separated. They are considered to be the pathological manifestation of peritoneal healing following surgery.[6,7] It is estimated that 93% of patients who undergo abdominal surgery will develop some postoperative adhesions.[8] In most cases, the adhesions do not translate into clinical symptoms; however, they can lead to serious complications, such as ASBO.

It is estimated that more than 300 000 emergency surgeries to treat ASBO are performed in the United States every year.[4] In the United Kingdom, small bowel obstruction is the indication for 37.3% of emergency laparotomies.[5] For patients who require hospital admission, the average hospital stay is 7 days, and the in-hospital mortality rate is 3%.[1] Small bowel obstruction also has a high recurrence rate.[9] Additionally, ASBO has been associated with a high financial burden. A study conducted in the Netherlands from 2013 to 2015 estimated the average cost of hospital admission and surgical treatment at €16 000.[10]

The management of ASBO remains a challenge. Peritonitis, strangulation, or bowel ischemia are the typical indications for urgent surgery.[11,12] The treatment of patients depends on the clinical judgment of the surgeon[13] and typically consists of intravenous fluid rehydration, fasting, and nasogastric tube decompression.[13,14]

Diatrizoate meglumine and diatrizoate sodium (Gastrografin) is a water-soluble contrast agent that can be used for visualization by X-ray or CT scan, and thus serves an important diagnostic function. However, Gastrografin can be helpful as a therapeutic agent as well. Due to a strong osmotic effect, it causes a shift of water into the lumen of the bowel, thereby facilitating the passage of stool, reducing edema of the intestinal wall, and helping resolve intestinal obstruction.[15] This dual function has been proven in multiple studies, systematic reviews, and meta-analyses.[15-23] As a diagnostic agent, the appearance of Gastrografin on an abdominal radiograph within 24 hours of administration is highly predictive of nonoperative resolution of obstruction.[16] As a therapeutic agent, Gastrografin reduces the time to resolution of the obstruction[15,17,18,21,22] and reduces the need for surgery.[17,18,20,22-24] Furthermore, Gastrografin has been proven safe. It does not increase morbidity or mortality.[16-18]

The diagnostic and therapeutic value of Gastrografin shortens the length of hospital stay,[15-23] which in turn leads to improvements in hospital efficiency, health care utilization, quality of health care, and patient satisfaction. Standard use of Gastrografin in ASBO patients has resulted in a tenfold cost reduction in overall inpatient care.[19] As a result, since 2013, the World Society of Emergency Surgery has recommended the routine use of Gastrografin for the diagnosis of ASBO and as part of the nonoperative treatment.[13]

In June 2019, a standardized protocol using Gastrografin for the management of patients with ASBO was implemented at Vancouver General Hospital (VGH) [Figure 1 [12]]. The aim of our study was to determine whether this protocol has improved the quality of patient care. Improved quality of care has been defined as shortened length of hospital stay, increased rate of successful conservative management, reduced need for surgery, shortened time to resolution of the obstruction, shortened time to surgery, reduced surgical complications, and patient satisfaction.

Methods

We conducted a single-centre, observational, nonrandomized controlled study with historical controls. The study was approved by the University of British Columbia Clinical Research Ethics Board and the Vancouver Coastal Health Research Institute. Two groups were studied: a retrospective preimplementation group (i.e., prior to implementation of the protocol) and a prospective postimplementation group managed with the standardized protocol. The preimplementation group was used as the historical control.

We included patients 18 years of age or older who had a primary diagnosis of ASBO. This was defined as adhesions being the most likely cause of obstruction based on the final CT scan report by the attending radiologist. Patients admitted to general surgery from November 2017 until October 2018 were included in the preimplementation group; those admitted from July 2019 until November 2019 were included in the postimplementation group.

We excluded patients who were in need of immediate surgery (peritonitis, strangulation, bowel ischemia, and closed-loop obstruction) based on clinical signs and CT scans. We also excluded those who were treated with surgery initially without a trial of conservative management. Patients with possible causes of a small bowel obstruction other than adhesions seen on CT were also excluded.

For the preimplementation group, management was dependent on the clinical judgment of the surgical team on call. In the postimplementation group, the Gastrografin protocol for ASBO was followed if the patient met the inclusion criteria.

Patient data were collected prospectively and were de-identified and stored in password-protected and encrypted computers. Baseline characteristics consisted of age, gender, and Charlson Comorbidity Index (CCI). We calculated the CCI value for each patient to categorize comorbidities. This is a combined age–comorbidity index that estimates 10-year survival.[25] The primary outcome measured was length of hospital stay, defined as the number of days from the patient’s admission date to their discharge date. The secondary outcome measures were rate of successful conservative management; need for surgery; time to resolution of the obstruction; time to surgery; readmission to hospital; rate of surgical complications, aspiration, or mortality; and patient satisfaction. Multivariate linear regression models were created to estimate the mean difference in continuous variables between the two groups, while adjusting for age, gender, and CCI. The effect measure applies additively as an adjusted mean difference. Multivariate logistic regression models were created to estimate the multiplicative odds ratios of the categorical variables, while adjusting for age, gender, and CCI. Statistical significance was defined as a P value < 0.05. Although the models were all multivariate, only the results pertaining to the effect of postimplementation or preimplementation group were selected. We used RStudio and Microsoft Excel for our analyses.

To gather feedback prospectively on the quality of care patients in the postimplementation group received during their stay in hospital for ASBO treatment, the validated Canadian Patient Experiences Survey on inpatient care was used.[26] The patients completed the survey on paper prior to discharge or over the phone after discharge. Ethical consent was obtained prior to administering the survey.

Results

The study included 122 patients: 82 in the preimplementation group and 40 in the postimplementation group. The mean (SD) age was 68 (16) years, and 59.0% of the patients were female. There were no statistical differences in the baseline characteristics between the two groups.

Primary outcome

Overall, the postimplementation group had a significantly shorter mean length of hospital stay than the preimplementation group (4.15 days versus 7.22 days; P = 0.007). The result remained significant after adjusting for age, gender, and CCI: length of stay of patients in the postimplementation group was on average 3.21 days shorter than that of patients in the preimplementation group (95% CI, −5.722 to −0.645; P = 0.015).

Secondary outcomes

Successful conservative management

More patients were treated successfully with conservative management in the postimplementation group than in the preimplementation group, but the difference was not statistically significant (90.0% versus 74.7%; P = 0.083). However, after adjusting for age, gender, and CCI, patients in the postimplementation group had significantly higher odds of being treated successfully with conservative management compared to patients in the preimplementation group (odds ratio: 3.354; 95% CI, 1.129 to 12.600; P = 0.044).

Need for surgery

Fewer patients in the postimplementation group required surgery compared to patients in the preimplementation group, but the difference was not statistically significant (7.5% versus 22.9%; P = 0.066). However, after adjusting for age, gender, and CCI, patients in the postimplementation group had significantly lower odds of needing surgery compared to patients in the preimplementation group (odds ratio: 0.237; 95% CI, 1.129 to 12.600; P = 0.034).

Time to resolution of the obstruction

The mean time from admission to resolution of the obstruction was significantly shortened for patients in the postimplementation group who were treated successfully with conservative management (1.74 days versus 2.77 days; P = 0.019). However, the difference was not significant after adjusting for age, gender, and CCI (adjusted mean difference: −1.059; 95% CI, −2.164 to −0.919; P = 0.06).

Time to surgery

There was no statistically significant difference in time to surgery between the two groups (3.17 days versus 3.66 days; P = 0.688). This persisted after adjusting for age, gender, and CCI (adjusted mean difference: −0.104 days; 95% CI, −4.547 to 0.910; P = 0.961).

Readmission to hospital

Fewer patients in the postimplementation group than in the preimplementation group needed readmission to hospital within 30 days of initial discharge, but the difference was not statistically significant (3.1% versus 10.1%; P = 0.433). There also was no significant difference between the two groups in the odds of being readmitted (odds ratio: 0.320; 95% CI, 0.017 to 1.909; P = 0.295).

Complications and mortality

Overall, there were no statistically significant differences in the rate of surgical complications, aspiration, or mortality between the two groups. No patients in the postimplementation group suffered from aspiration events. In the adjusted analysis, the odds ratios were not statistically significant.

Patient satisfaction

In the postimplementation group, 52.5% of patients (n = 21) rated their overall experience during their admission with a mean grade of 8.5 on a scale from 0 to 10, where 0 was the worst hospital experience possible and 10 was the best. The measure “hospital stay helpful” received a mean grade of 9.2 out of 10, where 0 was “not helped at all” and 10 was “helped completely.”

Patients generally found that they were treated with courtesy and respect by doctors, and that doctors listened carefully to them and explained things in a way they could understand. Seventy-five percent of the patients said they would recommend VGH to their friends and family [Figure 2 [13]].

Discussion

Length of hospital stay was significantly different between our two study groups: patients who received Gastrografin were discharged home 3 days sooner than patients who did not. This could be partially attributed to the higher odds of patients in the postimplementation group receiving successful nonoperative management. Our finding is in line with results reported in the literature.[15-23] The most significant difference in length of hospital stay between a control group and study group treated with Gastrografin was recorded in an open-label randomized controlled trial by Di Saverio and colleagues (7.8 days versus 4.7 days, respectively; P < 0.05).[22] By contrast, Scotté and colleagues suggested that no benefit was gained from Gastrografin administration in patients with ASBO in terms of length of hospital stay and need for surgery.[27] However, the authors set a cutoff period of 48 hours for the decision to perform surgery, as opposed to the 72-hour period recommended by the World Society of Emergency Surgery.[13] We believe that the earlier cutoff point may have resulted in a higher rate of surgical management, which in turn could have resulted in an increased length of hospital stay in patients who received Gastrografin. Additionally, Scotté and colleagues did not use a standardized protocol for Gastrografin.

The reduced length of hospital stay in the postimplementation group in our study may also have been due to the shortened time to resolution: patients in the postimplementation group had their nasogastric tube removed 1 day earlier than those in the preimplementation group. However, this difference was not significant after adjusting for age, gender, and CCI, possibly because the sample size of the subgroup of patients who did not undergo surgery was small.

The level of satisfaction with the quality of health care received by patients with ASBO has not been studied extensively in the literature. We measured this by using the Canadian Patient Experiences Survey on inpatient care.[26] Previous studies that used similar surveys have shown that surgical patients had a significantly higher satisfaction rate than medical patients.[28] Our results could be useful as a baseline or control for future studies in quality improvement that use our tailored patient survey, with questions specifically related to the patients’ condition in hospital and the treatment they received.

A major strength of our study is our demonstration of the value of using a protocol-based approach. To our knowledge, none of the published randomized controlled trials and systematic reviews on the use of Gastrografin had such a standardized protocol. Zielinski and colleagues conducted a prospective cohort study that compared the outcomes of patients managed on- and off-protocol using Gastrografin;[20] however, their protocol was optional and based on surgeon discretion. Another prospective cohort study by Weiss and colleagues also had a protocol, but it was initiated in only a portion of their patient population, not universally.[19] Long and colleagues conducted a more recent study that included a protocol using Gastrografin,[23] but the study was based on a retrospective review only.

Study limitations and recommendations

The sample size of our postimplementation group was small, mainly because of a time limitation in our data collection. This could explain why we did not find statistically significant differences between the two groups for all outcomes measured. Despite the limited sample size, our results are promising.

We recommend that a shared multi-institutional data registry be created so more data on more outcomes can be collected. Also, feedback about the protocol and its implementation should be gathered from the multidisciplinary surgical team so another survey can be developed and targeted toward the nursing staff.

Conclusions

An institutional standardized protocol using Gastrografin in the management of adhesive small bowel obstruction was associated with improved quality of patient care at Vancouver General Hospital. Most importantly, patients who received Gastrografin had a shorter hospital stay and were less likely to undergo surgical intervention. Patients were satisfied with the care they received. This study provided valuable benchmark data for further multi-institutional research. However, variability in the length of hospital stay, time to resolution of obstruction, and time to surgery should be further explored to identify the potential for improvement in patient care.

Competing interests

None declared.

This article has been peer reviewed.

Dr Zhou is a PhD candidate in the Faculty of Medicine at Erasmus University Rotterdam in the Netherlands. Mr Udwadia is a medical student in the Faculty of Medicine at the University of British Columbia. Mr Chen is a biostatistician in the Department of Surgery at the University of British Columbia. Dr Joos is a trauma surgeon and clinical assistant professor in the Department of Surgery at the University of British Columbia.