j o u r n a l o f s u r g i c a l r e s e a r c h  m a y 2 0 1 8 ( 2 2 5 ) 4 0 e4 4

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.JournalofSurgicalResearch.com

Relationship between duration of preoperative

symptoms and postoperative ileus for small
bowel obstruction

Whitney S. Brandt, MD,* Joshua Wood, BS,

Bishwajit Bhattacharya, MD, FACS, Kevin Pei, MD, FACS,
Kimberly A. Davis, MD, MBA, FACS, FCCM,
and Kevin Schuster, MD, MPH, FACS, FCCM
Department of Surgery and New Haven, Sections of General Surgery, Trauma and Surgical Critical Care,
Yale School of Medicine, New Haven, Connecticut

article info abstract

Article history: Background: Factors associated with postoperative ileus and increased resource utilization
Received 6 September 2017 for patients who undergo operative intervention for small-bowel obstruction are not
Received in revised form extensively studied. We evaluated the association between total duration of preoperative
20 November 2017 symptoms and postoperative outcomes in this population.
Accepted 19 December 2017 Materials and methods: We performed a retrospective review of patients who underwent
Available online 30 January 2018 surgery for small-bowel obstruction (2013-2016). Clinical data were recorded. Total dura-
tion of preoperative symptoms included all symptoms before operation, including those
Keywords: before presentation. Primary endpoint was time to tolerance of diet. Secondary endpoints
Small-bowel obstruction (SBO) included length of stay, total parenteral nutrition use, and intensive care unit admission.
Postoperative ileus Association between variables and outcomes was analyzed using univariable analysis,
Adhesiolysis multivariable Poisson modeling, and t-test to compare groups.
Length of stay Results: Sixty-seven patients were included. On presentation, the median duration of
Utilization of resources symptoms before hospitalization was 2 d (range 0-18 d). Total duration of preoperative
symptoms was associated with time to tolerance of diet on univariable analysis (Pearson’s
moment correlation: 0.28, 95% confidence interval: 0.028-0.5, P ¼ 0.03). On multivariable
analysis, ascites was correlated with time to tolerance of diet (P < 0.01), but total duration
of preoperative symptoms (P ¼ 0.07) was not. Length of stay (Pearson’s correlation: 0.24,
95% confidence interval: 0.02 to 0.47, P ¼ 0.07) was not statistically different in patients
with longer preoperative symptoms. Symptom duration was not statistically associated
with intensive care unit (P ¼ 0.18) or total parenteral nutrition (P ¼ 0.3) utilization.
Conclusions: Our findings demonstrate that preoperative ascites correlated with increased
time to tolerance of diet, and duration of preoperative symptoms may be related to post-
operative ileus.
ª 2017 Elsevier Inc. All rights reserved.

* Corresponding author. General Surgery, Trauma and Surgical Critical Care, Yale School of Medicine, 330 Cedar Street, BB 310, PO Box
208062, New Haven, CT 06520. Tel.: þ(203) 785 2572; fax: þ(203) 785 3950.
E-mail address: whitney.brandt@yale.edu (W.S. Brandt).
0022-4804/$ e see front matter ª 2017 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.jss.2017.12.031
 brandt et al  preop symptoms and postop ileus in sbo 41

Introduction signs of bowel ischemia or clinical decompensation. The de-

cision to operate electively was based on surgeon discretion.
Small-bowel obstructions (SBO) account for almost 300,000
hospital admissions per year.1-4 Previous studies have demon- Statistical analysis
strated that the average cost per admission for SBO ranges from
$30,000-$38,000, which given the approximate number of ad- Variables were analyzed via bivariate and multivariable
missions per year, is estimated to cost 9-11.4 billion dollars per techniques. Pearson’s productemoment correlation was used
year.5 Current guidelines recommend initial nonoperative
management for patients without generalized peritonitis or
evidence of clinical deterioration.6 Only 30% of patients who Table 1 e Patient demographics.
present with SBO require immediate operative intervention.
Characteristic (n ¼ 67) Median or N Range
However, of the 70% of patients who undergo initial nonopera-
or %
tive management, 23%-36% do not resolve and eventually
require operative intervention.5 Given the large amount of pa- Age (y) 68 20-95

tients who undergo operative intervention for SBO, factors Sex

associated with longer postoperative ileus and higher utilization Male 31 46%
of resources would be helpful to define high-risk populations. Female 36 54%
Surprisingly, few studies have been performed which Comorbidities
evaluate factors associated with prolonged ileus and increased
Diabetes mellitus 11 17%
resource utilization in patients who undergo operative inter-
Hypertension 19 29%
vention for SBO. One simple factor which may be associated
Coronary artery disease 5 7.6%
with poor outcomes in this population is total duration of
preoperative symptoms. In our study, we aim to evaluate the Chronic kidney disease 8 12%

association between the total duration of preoperative symp- Previous malignancy 5 7.6%
toms, including symptoms before hospitalization on post- Previous cerebrovascular accident 6 9.1%
operative ileus, length of stay (LOS), intensive care unit (ICU) Thyroid disease 5 7.6%
admission, and total parenteral nutrition (TPN) use. Liver disease 1 1.5%
Timing of operation
Emergent 40 60%
Methods
Non-emergent 27 40%
Operative duration (min) 146 53-354
After approval by the Institutional Review Board, we per-
formed a retrospective review of consecutive patients older Cause of bowel obstruction
than 18 y who underwent operative intervention for me- Adhesions 50 74%
chanical bowel obstruction (i.e., adhesive, intussusception, Internal hernia 11 16%
hernia, malignancy, and so forth) between January 2013 and Abdominal wall hernia 8 12%
June 2016. Patients with isolated large-bowel obstruction, no Closed loop obstruction 13 19%
bowel obstruction at the time of surgery, or large-bowel
Volvulus 4 6%
involvement with SBO were excluded from the study. Pa-
Small-bowel mass 3 4%
tients with two operative interventions for SBO were only
Other 4 6%
included once for analysis.
Patient demographics, duration of symptoms, nasogastric Bowel ischemia in operating room 18 27%
tube (NGT) output in the 24 h before surgery, operative details, Ascites in operating room 22 33%
radiographic findings, and postoperative clinical data were Operative approach
collected by chart review. Total duration of preoperative Laparoscopy 14 21%
symptoms was defined as duration of prehospitalization Laparotomy 53 79%
symptoms plus duration of symptoms after admission before
Time to tolerance of diet (d) (n ¼ 60) 5 0-21
operation in days. Nasogastric tube output in the 24 h before
Length of stay (d)
operation was recorded for patients with nasogastric tubes.
Total length of stay 9 3-75
The time to tolerance of diet was defined as the time after
surgery to tolerance of input without further nausea or vom- Postoperative length of stay 7 1-74

iting. Length of stay (LOS) was time from admission to Postoperative resource utilization
discharge in days. Operative time was defined as start of Total parenteral nutrition 15 22%
incision to time of closing incision (minutes; continuous). Intensive care unit admission 23 34%
Emergent operative intervention was defined as the decision Disposition
to operate based on clinical signs of decompensation (i.e.,
Death 6 9.0%
peritonitis, unstable vitals) and/or radiographic signs of bowel
Home without services 29 43%
ischemia. All other cases were classified as planned operative
Services, rehabilitation, or hospice 16 24%
intervention, including any non-emergent surgery without
42 j o u r n a l o f s u r g i c a l r e s e a r c h  m a y 2 0 1 8 ( 2 2 5 ) 4 0 e4 4
to evaluate the association between continuous variables
including time to tolerance of diet and LOS. Chi-square test
was used to evaluate categorical variables. t-Test was used to
directly compare LOS or time to tolerance of diet between
groups. A multivariable analysis was performed by creating
Poisson models to evaluate the effect of total duration of
preoperative symptoms, presence of ascites in the operating
room, need for bowel resection, and operative time on post-
operative time to tolerance of diet. The primary endpoint was
postoperative time to tolerance of diet. Secondary endpoints
included LOS, postoperative ICU admission, and use of TPN.
The level of statistical significance was set to P < 0.05.
Results
Patient population
Out of the 82 consecutive patients originally included, 15 pa-
tients were found to have large-bowel obstruction at surgery
or no mechanical bowel obstruction and were excluded from
the study. A total of 67 patients were included in the analysis.
The median age was 68 y with a range from 20-95 y. Forty-six
percent of the patients were male, and 54% of the patients
were female. Patient comorbidities, demographics, and oper-
ative and clinical details are given in Table 1.
Patient presentation
On presentation, the median duration of symptoms before
hospitalization was 2 d (range 0-18 d). For all patients (n ¼ 67),
the median time from evaluation to incision was 18 h (range
1-678 h). Sixty-one percent (n ¼ 41/67) of patients underwent
operative intervention within 48 h. Of the patients who un-
derwent operative intervention after 48 h, the median time
after admission to operation was 3.5 d (range 2-42 d).
Thirty-three patients (49%) had a NGT in the 24 h before
operation. The median NGT output in the 24 h before the oper-
ating room was 800 cc (range 100-3720). All patients had preop-
erative computed tomography (CT) scan. The radiographic
report for the CT scan indicated presence of ascites in 49 of 67
patients. No patients underwent small-bowel follow-through.
Operating room details
In total, 40 patients underwent emergent operative interven-
tion. Upon exploration, 45% (n ¼ 30) patients were found to
Table 2 e Bivariate analysis of time to tolerance of diet and length of stay.
Variable Pearson’s product estimated correlation (95% confidence interval)
Time to tolerance of diet
Time after presentation to operation (h)
Total duration of preoperative symptoms (d)
NGT output 24 h before operation (cc)
Length of stay (LOS)
Time after presentation to operation (h)
Total duration of preoperative symptoms (d)
have isolated adhesions as origin of their bowel obstruction
(Table 1). The median duration of operation was 146 min
(range 53-354 min).
Of the population, 46% (n ¼ 31) underwent bowel resection.
Eighteen patients (27%) had bowel ischemia necessitating
resection, 9 patients (13%) had resection secondary to iatro-
genic injury, and in the remaining 6 patients, the reason for
resection was undocumented. Only 4 patients (6%) had a
perforation at the time of operation, and 24 patients (36%) had
ascites noted in the operating room report.
Utilization of resources
We first analyzed how total duration of preoperative symp-
toms was related to postoperative time to diet tolerance by
evaluating the correlation between multiple preoperative
variables, including total duration of preoperative symptoms
to postoperative time to tolerance of diet (Table 2). We then
found that total duration of preoperative symptoms (in days)
was correlated with time to tolerance of diet on univariable
analysis (Table 2). After demonstrating an association on
univariable analysis, we then performed a multivariable
analysis using Poisson modeling and found that only ascites
was statistically, significantly associated with postoperative
time to tolerance of diet (95% confidence interval [CI]:
0.46-0.86, P  0.01). Total duration of preoperative symptoms,
need for bowel resection, and operative time were not statis-
tically, significantly associated with time to diet tolerance on
multivariable analysis (Table 3).
Because we had some evidence to support that total
duration of preoperative symptoms was associated with time
to tolerance of diet, we were interested in whether total
duration of symptoms was associated with longer LOS. How-
ever, we did not find a statistically significant association
between duration of preoperative symptoms and LOS
(Table 2). In addition, while there was a numerical difference
in mean LOS between patients who were taken directly to the
operating room at time of presentation (mean LOS: 4.73 d)
versus those who underwent initial nonoperative manage-
ment (mean LOS: 6.88 d) this difference was not statistically
significant (t-test, P ¼ 0.06).
Finally, we were interested in whether total duration of
preoperative symptoms was associated with utilization of
other resources. We found that, although patients who
required TPN or ICU admission had a higher average total
duration of preoperative symptoms, these differences were
not statistically significant (Figure A and B).
P value
0.24 (0.02 to 0.47) 0.07
0.28 (0.028 to 0.5) 0.03
0.01 (0.40 to 0.38) 0.95
0.12 (0.13 to 0.35) 0.35
0.05 (0.19 to 0.29) 0.68
 brandt et al  preop symptoms and postop ileus in sbo
Table 3 e Multivariable analysis for effect of predictors on time to tolerance of diet.
Variable
Total preoperative duration of symptoms (d)
Ascites
Need for bowel resection
Operative time (min)
*
Increase in time to tolerance of diet in days per unit increase in the predictor variable.
Discussion
Given that there are almost 4,000,000 admissions for SBO with
a cost ranging from $29,000 to $86,000 per admission, it is
Fig e Differences in total duration of preoperative
symptoms by resource. (A) Total duration of preoperative
symptoms by ICU admission. Patients admitted to the ICU
(n [ 23) had an average duration of preoperative
symptoms of 12 d (95% CI: 1.7-12.11) versus 3.7 d (95% CI:
2.6-4.8) for patients who were not admitted to the ICU
(n [ 44). (B) Total duration of preoperative symptoms by
TPN administration. Patients who required TPN (n [ 15)
had an average total duration of preoperative symptoms of
8.5 d (95% CI: 0.3-1.6 d) versus 3.8 d (95% CI: 2.8-4.8 d) for
patients who never required TPN (n [ 52).
43
Model coefficients* (95% confidence interval) P value
1.00 (0.99-1.00) 0.07
0.63 (0.46-0.86) <0.01
1.35 (0.91-2.00) 0.13
1.00 (1.00-1.01) 0.05
important to find factors associated with higher utilization of
resources in patients who undergo operative intervention for
SBO.4,7 Here, we aimed to evaluate whether total duration of
preoperative symptoms may be a factor which is associated
with poor postoperative outcomes, including increased utili-
zation of hospital resources.
On univariable analysis, we found that total duration of
symptoms was associated with increased postoperative
ileus. The association between longer duration of symp-
toms and longer postoperative ileus may be secondary to a
longer duration of bowel dilation, and therefore, a longer
time for the bowel to return to its normal physiologic size
and state.
Interestingly, while we found that total duration of pre-
operative symptoms correlates to time to tolerance of diet on
univariable analysis, there was no association between total
duration of preoperative symptoms and postoperative LOS
(P ¼ 0.68). This lack of difference may be skewed secondary to
the socioeconomic status of our patient population and the
associated difficulty establishing a destination for disposition.
Conversely, the total LOS in patients who were taken directly
to the operating room upon presentation versus those who
initially underwent nonoperative management approached
statistically, significantly shorter lengths of stay (4.73 versus
6.88 d, P ¼ 0.06). This suggests that earlier intervention among
those requiring operative intervention may be associated with
a shorter LOS. Our study may have been underpowered to
determine statistical significance.
We also evaluated whether there was a difference in
average duration of preoperative symptoms in patients who
required TPN or ICU admission. Although patients who
required TPN or ICU admission had higher average total
duration of preoperative symptoms, these differences were
not statistically different. It is possible that patients with
longer duration of preoperative symptoms have larger dilation
of the bowel, and therefore, will need more resources while
their bowel returns to normal physiologic state. Further
studies to evaluate this idea with a larger subset of patients
may be helpful to determine whether this association truly
exists.
Given that high nasogastric tube output may be associated
with a higher degree of obstruction and a longer eventual time
to recovery, we evaluated whether high nasogastric tube
output may be one such predictor of postoperative ileus.
Interestingly, we found no association between nasogastric
tube output in the 24 h before the operating room with post-
operative time to tolerance of diet (P ¼ 0.95). This lack of dif-
ference may be secondary to a selection bias for placement of
44 j o u r n a l o f s u r g i c a l r e s e a r c h  m a y 2 0 1 8 ( 2 2 5 ) 4 0 e4 4
nasogastric tubes in patients with more proximal or symp-
tomatic obstructions. In addition, depth of nasogastric tube
placement was not controlled. More distally placed tubes may
have falsely elevated outputs, which could contribute to the
lack of correlation. All these factors may explain the lack of
correlation between nasogastric tube output and post-
operative time to tolerance of diet.
Multivariable Poisson modeling was used to delineate
percent change of time to tolerance of diet per unit increase
in predictor variable. Interestingly, only ascites was found to
have a statistically significant association with time to
tolerance of diet (P < 0.01). This finding seems counterintu-
itive, as in that ascites is usually associated with bowel
ischemia. We therefore postulate that perhaps ascites was a
marker of bowel ischemia and that bowel ascites on imaging
may lead to selection bias leading to earlier operative
intervention and therefore shorter postoperative ileus in this
subset of patients. In our population, 49 of 67 patients were
found to have ascites on CT scan before operation. Of the
patients with preoperative CT scan ascites, the average
duration from evaluation until operative intervention was
37.5 h versus 85 h in patients who did not have CT scan
ascites (P ¼ 0.04). This indicates that likely there is some
selection bias for earlier operative intervention in patients
with CT scan ascites which may have lead to decreased
postoperative ileus in this population.
We also evaluated whether a bowel resection, which may
indicate the presence of bowel ischemia, was associated with
postoperative ileus. However, we found no association be-
tween the need for bowel resection and postoperative time to
tolerance of diet (95% CI: 0.91-2.00, P ¼ 0.13) or operative time
(95% CI: 1.00-1.01, P ¼ 0.052). It is unclear if the lack of corre-
lation between operative time and postoperative time to
tolerance of diet is related to lack of power, surgeon variability
in operative technique, or a true lack of association.
There are a few limitations to our study. For one, this is a
retrospective single-institution study and therefore may not
be generalizable to all patient populations. In addition, our
sample size was relatively small and our population had
significant heterogeneity. A larger cohort would likely reduce
this heterogeneity and add statistical power. Another limi-
tation involves differences in postoperative management
from various surgeons. For example, differences in post-
operative diet advancement by various surgeons may have
led to variability in time to tolerance of diet. In addition, for
those patients who underwent initial nonoperative man-
agement, decision to operate was as per surgeon discretion.
While this may be a limitation due to lack of defined pro-
tocol, we also believe this to be a strength as it is reflective of
the true heterogeneity in management of SBO. Finally,
although duration of preoperative symptoms would be a
helpful marker for postoperative ileus, it has potential for
recall bias. A more objective marker for degree of obstruction
may be preobstruction bowel caliber, which may be an
additional helpful surrogate for degree of bowel obstruction
at presentation.
Conclusions
Overall, our study demonstrated that total duration of pre-
operative symptoms may be associated with a longer post-
operative ileus, potential for higher utilization of hospital
resources, including a longer LOS, and potentially increased
likelihood to require ICU admission. Consideration of total
duration of symptoms may help predict return of bowel
function, aid in counseling patients, and anticipating resource
utilization.
Acknowledgment
Authors’ contributions: W.S.B. contributed to data gathering,
original manuscript drafting, writing, reviewing, and editing.
J.W. was also a part of data gathering, writing, reviewing, and
editing. B.B. and K.A.D. contributed to conceptual design,
writing, reviewing, editing, and supervision. K.P. was involved
in writing, reviewing, and editing. K.S. was also responsible for
conceptual design, writing, reviewing, editing statistical
evaluation, original manuscript draft, review, editing, and
supervision.
Disclosure
The authors reported no proprietary or commercial interest in
any product mentioned or concept discussed in this article.
references
1. Sikirica V, Bapat B, Candrilli SD, Davis KL, Wilson M, Johns A.
The inpatient burden of abdominal and gynecological
adhesiolysis in the US. BMC Surg. 2011;11:13.
2. Scott FI, Osterman MT, Mahmoud NN, Lewis JD. Secular trends
in small-bowel obstruction and adhesiolysis in the United
States: 1988-2007. Am J Surg. 2012;204:315e320.
3. Ray NF, Denton WG, Thamer M, Henderson SC, Perry S.
Abdominal adhesiolysis: inpatient care and expenditures in
the United States in 1994. J Am Coll Surg. 1998;186:1e9.
4. Jafari MD, Jafari F, Foe-Paker JE, et al. Adhesive small bowel
obstruction in the United States: has Laparoscopy made an
Impact? Am Surg. 2015;81:1028e1033.
5. Bilderback PA, Massman 3rd JD, Smith RK, La Selva D,
Helton WS. Small bowel obstruction is a surgical disease:
patients with adhesive small bowel obstruction requiring
operation have more cost-effective care when admitted to a
surgical service. J Am Coll Surg. 2015;221:7e13.
6. Maung AA, Johnson DC, Piper GL, et al. Evaluation and
management of small-bowel obstruction: an Eastern
Association for the Surgery of Trauma practice management
guideline. J Trauma Acute Care Surg. 2012;73(5 Suppl
4):S362eS369.
7. Oyasiji T, Angelo S, Kyriakides TC, Helton SW. Small bowel
obstruction: outcome and cost implications of admitting
service. Am Surg. 2010;76:687e691.