Abstract
Abstract
Background:
Approximately one-third of children with appendicitis present with advanced disease or perforation. Whereas this increases the risk for post-operative complications and re-admission, it is not yet possible to predict early on who will develop an abdominal abscess. We sought to identify specific risk factors for this complication, in an attempt to streamline post-operative care.
Patients and Methods:
We reviewed the records of all cases of perforated appendicitis over a 12-month period at a tertiary children's hospital. All patients who developed an abscess despite treatment minimum of seven days of antibiotic therapy were identified. Patients who presented or were re-admitted with an abscess were excluded from analysis. Records were reviewed for demographics, laboratory results, progression of oral intake, and vital signs.
Results:
Of 273 patients with appendicitis, we identified 59 cases of perforated appendicitis. Fifteen patients were excluded. Eight of the remaining 44 patients (18.2%) developed an abscess during their initial admission. Their mean length of stay was longer than that of patients without an abscess (13.4 ± 7.1 vs. 6.9 ± 1.9 d, p < 0.0001). Gender, leukocytosis, or diarrhea at presentation, maximum temperature on post-operative day 3, and maximum heart rate on post-operative day 3 were not statistically different. Diet progression was different between the two groups: none of the 21 patients who were tolerating a regular diet by post-operative day 3 developed an abscess, compared with 8 of the 23 patients who were not yet eating a regular diet on post-operative day 3 (p < 0.01). Late leukocytosis also correlated with the presence of an abscess: 7 of the 8 patients with an abscess had persistent leukocytosis at days 5 through 7, compared with 3 of 31 patients without abscess (p < 0.05). An ultrasound was obtained for these 3 patients and proved normal.
Conclusions:
Tolerating a regular diet three days after appendectomy for perforated appendicitis decreased the likelihood of a post-operative abscess. No other parameter was predictive of this complication early in the post-operative period. If confirmed in a larger prospective study, this finding may help decrease the length of stay for low-risk patients, and identify abscesses in high-risk patients in a timely fashion.
A
In the current literature, risk factors at presentation for the formation of intra-abdominal abscess include diarrhea, age, weight, and body mass index [5,12]. At the time of surgery, an intra-operative appendicolith is associated with an increased incidence of intra-abdominal abscess development [12]. Post-operative risk factors that have been suggested include lymphocyte depression, fever on post-operative day 3, and an increased white blood cell count on post-operative day 5 [12,13]. However, diarrhea at presentation is the only risk factor that has been confirmed in more than one study [5,12] and Henry et al. [12] did not find increased maximum temperature on post-operative day 3 to be predictive, as did Fraser et al. [5].
Early prediction of adverse outcome in the post-operative course could help guide clinicians to optimize surgical care. Recent efforts have focused on creating clinical practice guidelines to aid in standardizing care and improving patient outcomes [14]. These guidelines can streamline care and identify patients at risk of developing a post-operative abscess, which helps plan for early imaging and minimizes re-admission. This study sought to identify demographic and in-hospital characteristics that place children with complicated appendicitis at increased risk of developing a post-operative intra-abdominal abscess.
Patients and Methods
Study design
This study is a retrospective case control study. We reviewed our records of patients with perforated appendicitis who went to the operating room over a 12-month period (2015–2016) at a tertiary children's hospital. Perforated appendicitis was defined surgically and confirmed pathologically (not radiographically) [15]. Patients were excluded if they presented with perforation and abscess already present on imaging, or were initially managed non-operatively or with vascular and interventional radiology (VIR) drainage. Type of appendectomy was left to surgeon's preference: we and others have shown previously that the incidence of post-operative abscesses is not influenced by the surgical approach [16]. All patients younger than 18 years of age who developed an abscess despite treatment consisting of a seven-day course of antibiotics were included. Patients re-admitted with an abscess after early discharge (considered incomplete antibiotic therapy) were excluded from analysis. Charts were reviewed for demographics, laboratory results over the course of stay, progression of diet, vital signs over the course of stay, diagnostic imaging, and operative details. Patients were considered to have developed a post-operative abscess based on radiographic diagnosis either with ultrasound or computed tomography (CT) scan. This study was approved by the Hasbro Children's Hospital/Rhode Island Hospital Institutional Review Board.
Statistical analysis
Patient demographics and characteristics of those who developed an in-house abscess were compared with those who did not develop an abscess. For categorical data, analyses were conducted using χ2 statistic, Fisher exact test, and a two-tailed p value. For continuous data of one variable between two groups, an unpaired t-test was used. A p ≤ 0.05 was considered significant for all analyses.
Results
We reviewed 273 cases of appendicitis and identified 59 cases of perforated appendicitis. Fifteen patients were excluded: six had an abscess identified at the time of presentation, three patients did not have a full course of antibiotics, one was treated only with antibiotics, and the remaining five did not meet the evidence-based definition of perforation. Therefore 44 patients with perforated appendicitis were included. Eleven of the 44 underwent open appendectomy, which was based on surgeon's preference. There was no difference in abscess development based on the operation performed (open 2/11 (18.2%) abscess rate vs. laparoscopic 6/33 (18.2%) abscess rate, p = 1.0). Eight of the 44 patients (18.2%) developed an abscess during their initial admission. Their mean length of stay was longer than that of patients without an abscess (13.4 vs. 6.9 d, p < 0.0001; Table 1).
SD = standard deviation; LOS = length of stay; WBC = white blood cell count.
None of the patient demographics or clinical findings at presentation (age, gender, leukocytosis, or diarrhea) was different between those patients who developed an abscess and those who did not. Post-appendectomy parameters, such as of maximum temperature on post-operative day 3 and maximum heart rate on post-operative day 3, were not statistically different. Of note, no patient who developed an abscess had a fever on post-operative day 3. Seven of the eight patients with an abscess had persistent leukocytosis (defined as white blood cell count greater than 12.0 × 109/L) at days 5 through 7, compared with 3 of 31 patients without abscess (Table 2; p < 0.001). An ultrasound was obtained in these three patients and proved normal.
bpm = beats per minute.
Diet progression was significantly different between the two groups: none of the 21 patients (0%, confidence interval [CI] 0–15.5%) who were tolerating a regular diet by post-operative day 3 developed an abscess, compared with 8 of the 23 patients who were not yet eating a regular diet on post-operative day 3. Five patients had a post-operative ileus defined as inability to tolerate a diet until post-operative day 7, none of whom developed a post-operative abscess.
Discussion
Of those patients who present with perforated appendicitis 15% to 25% will develop a post-operative deep space abscess or surgical site infection (SSI), despite appropriate antibiotic therapy (5–7 d) [16]. To date, no early peri-operative factors can predict reliably who will develop an abscess. Post-operative risk factors include lymphocyte depression (a rare event suggesting severe sepsis), fever on post-operative day 3 (a relatively sensitive but not specific parameter), and an increased white blood cell count on fever on post-operative day 5 [5,13]. Early imaging is generally discouraged, because post-operative changes and residual fluid pockets cannot be easily distinguished from collections that will develop into true abscesses.
Persistent fever or leukocytosis by post-operative days 5 through 7 raises the suspicion of an abscess, as shown in our study. Imaging is then typically recommended, leading to image-guided percutaneous drainage or operative re-intervention if an abscess is identified. This increases length of hospitalization by an average of 6.5 days (13.4 ± 7.1 d vs. 6.9 ± 1.9 d if no abscess is found). Whereas persistent leukocytosis on days 7 through 7 is helpful in raising suspicion for an abscess, it is a relatively late finding, and deciding to obtain further work-up at that time may prolong hospitalization needlessly. Conversely, discharge without diagnostic work-up increases the risk of re-admission for infectious complications. Both prolonged hospitalization and re-admission increase resource utilization and cost, decrease patient and family quality of life (health utility), and may, in the future, lead to decreased reimbursement by third-party payers [17].
Thus, earlier predictors (peri-operatively, or well before post-operative day 5) would allow earlier diagnostic work-up and intervention, therefore reducing hospital stay and minimizing the risk of re-admission. Absence of an abscess in patients at risk would allow timely discharge of those patients and, possibly, earlier completion of antibiotic therapy.
In this small retrospective study, none of the at-risk patients who were found (by post-operative days 6 through 7) to have an intra-abdominal abscess were tolerating a regular diet by post-operative day 3. Conversely, none of the patients who were tolerating a regular diet by post-operative day 3 were later found to have an abscess despite the fact that many (5/22, 22.7%) continued to have intermittent fevers past the 3-day mark. Leukocytosis was more present in patients with an abscess, but this factor was not foolproof: 1 of 8 patients with an abscess had a normal white blood cell count by day 7, whereas 3 of the 31 patients without abscess had leukocytosis.
A larger, prospective study is needed to support these findings in greater detail, and to determine how foolproof diet progression is in the early prediction of post-operative abscesses. At this time, we are developing best-practice guidelines, whereby patients with perforated appendicitis who are not tolerating a regular diet by post-operative day 3 undergo diagnostic imaging on post-operative day 5. If no abscess is found, these patients are discharged home on oral antibiotics. Our data support the potential safety of discharging patients earlier on a course of oral antibiotics if they are tolerating a diet by post-operative day 3 and afebrile, which has been recently demonstrated by Willis et al. [14]. Patients who are tolerating a regular diet by post-operative day 3 but have persistent fevers beyond that time similarly undergo diagnostic imaging on post-operative day 5. We believe that this targeted approach will allow earlier discharge without risk of re-admission in all patients with perforated appendicitis.
Footnotes
Acknowledgments
We would like to acknowledge the assistance of Debra Watson-Smith, BSN, RN, for her assistance with the Institutional Review Board approval process and support.
Author Disclosure Statement
No competing financial interests exist.