- Email: [email protected]
In-hospital perforation risk in acute appendicitis: Age matters
The American Journal of Surgery xxx (xxxx) xxx
Contents lists available at ScienceDirect
The American Journal of Surgery journal homepage: www.americanjournalofsurgery.com
In-hospital perforation risk in acute appendicitis: Age matters Keith A. Hanson, Daron Jacob, Adel Alhaj Saleh, Sharmila Dissanaike* Department of Surgery, Texas Tech University Health Sciences Center, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 March 2019 Received in revised form 20 May 2019 Accepted 28 May 2019
Background: Controversy exists regarding how quickly an adult with appendicitis requires surgery to prevent perforation, and recent literature on antibiotic use as definitive treatment has complicated this question further. Since perforation is associated with worse outcomes, particularly in the elderly, efforts to prevent this complication are warranted. We studied risk factors for in-hospital perforation in patients diagnosed by admission CT with non-perforated acute appendicitis. Methods: We evaluated baseline demographics, symptom duration, and time from admission to antibiotics and surgery. Outcome measure was perforation diagnosed intra-operatively by attending surgeon. Results: Of 700 patients, 84 (12%) sustained in-hospital perforation; time from admission to operation or antibiotics were not associated. Duration of symptoms >24 h (aOR ¼ 2.23, 95% CI ¼ 1.33e3.72, p < 0.001) increased perforation risk. Patient age over 46 years (aOR ¼ 4.54, 95% CI ¼ 2.04e10.06, p < 0.001) was also associated with higher risk that increased with increasing age. Conclusion: Time to operation and antibiotic timing were not associated with in-hospital perforation in a general adult population. However, these findings suggest a possible benefit to expedient surgery in older patients. © 2019 Elsevier Inc. All rights reserved.
Introduction Acute appendicitis is one of the most common surgical emergencies worldwide.1 Perforation of the appendix, which increases the risk of abscess formation, wound infection, and sepsis, is a major source of morbidity associated with the condition.2,3 Perforation also increases mortality, with the risk of death varying from double the risk in young, otherwise healthy individuals to eight times the risk in the elderly when compared to uncomplicated appendicitis.4 While appendectomy is generally accepted as the treatment of choice, controversy exists regarding how quickly an adult with appendicitis requires surgery to prevent perforation and post-operative complications. The recent literature on antibiotic use as definitive treatment for appendicitis has complicated this question further, since it is theoretically feasible that antibiotic administration prevents perforation in appendicitis.5e10 Our objective in this study was to determine which variables influenced in-hospital perforation in adults who initially presented to the emergency room with uncomplicated, acute appendicitis as confirmed by CT scan. We hypothesized that a longer time from
* Corresponding author. 3601 4th St, Lubbock, TX, 79424, USA. E-mail address: [email protected] (S. Dissanaike).
admission to surgery would be associated with an increased incidence of perforation in patients with an initial diagnosis of nonperforated appendicitis. Additionally, we wanted to test whether timing of antibiotic administration affected the occurrence of inhospital perforation in these patients. Methods A retrospective chart review was conducted at an academic teaching hospital of patients who underwent appendectomy with a presumptive diagnosis of acute, non-perforated appendicitis, between January 2012 and December 2017. After IRB approval was obtained, ICD-9 code 540.9 (acute appendicitis) was used to procure a list of patient records for the given time frame. Patients who were found to have a perforated appendix on their initial CT scan were excluded from the study. Pregnant women and children were also excluded from the study, and subjects were required to have been between 18 and 89 years of age at the time of the operation. Transfer patients were included if adequate documentation was available from the outside facility. The majority of transfers were from local urgent care centers, freestanding emergency rooms and rural hospitals within a 100 mile radius. Data collected included perforation status, segmented neutrophil count, WBC count, duration of symptoms, sex, ethnicity, timing from antibiotic
https://doi.org/10.1016/j.amjsurg.2019.05.015 0002-9610/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
2
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
administration to OR, and timing from admission to OR. Regular treatment protocol for appendicitis at this institution is for surgery to be consulted immediately on any patient with suspected appendicitis. Most patients receive a CT scan to confirm the diagnosis by the ER physician prior to consult. Dedicated in-house emergency general surgeons are available 24/7/365, therefore there is no delay in treatment due to conflict with elective surgery schedules; however, there may be delay due to more emergent cases. The practice of the surgical group at our institution is to offer laparoscopic appendectomy to all patients diagnosed with acute appendicitis. In order to obtain true informed consent, patients are also informed of the option of non-operative management with antibiotics, however the standard surgeon recommendation is for operative treatment. Occasionally a trial of non-operative management may be offered at attending surgeon discretion after extensive discussion with the patient in cases of true clinical equipoise as to the diagnosis; this is the exception, not the norm. Abdominal CT scans with or without IV contrast were reported by experienced, board-certified radiologists, and these reports were used as the definitive preoperative diagnosis (perforated vs. non-perforated appendicitis). The radiological criteria for perforation on CT was defined as the criteria for appendicitis in addition to one or more of the following: intra-abdominal abscess, extraluminal air, or extra-luminal appendicolith. If perforation was explicitly stated or reported that it could not be ruled out on the radiologist report of the CT, that patient was excluded from the study. If the radiologist's report made no mention of the possibility of perforation, then the patient was considered non-perforated on admission, and included as a subject in the study. The determination of whether a subject experienced perforation between hospital admission and OR was based on OR findings described in the operative notes from the board-certified attending surgeons performing or supervising the appendectomy. Admission time was based on the emergency department nursing note, and all data were collected using electronic medical records. Differences between the perforated and unperforated group were determined using the Student t-test and Wilcoxon rank sum test as appropriate for white blood cell counts and segmented neutrophil counts. The relationship between WBC, neutrophil counts and perforation status were too weak to justify further analysis by value categories with establishment of cutoff values. Chi-square or Fishers exact test were used for the variables race, sex, age, time from admission to OR, time from antibiotic administration to OR, and duration of symptoms. These variables were divided into quartiles and analyzed by category both to maintain as much statistical rigor as possible and to better establish cutoffs that would be more clinically relevant. Reported differences between transfer and non-transfer patients were determined using the Student t-test and Wilcoxon rank sum test as appropriate for all variables. Unadjusted logistic regression analysis was conducted to determine the association between individual predictors and inhospital perforation. Based on the maximum likelihood estimates, only those variables that showed significant associations (p < 0.05) with the outcome were considered for the adjusted logistic regression model. All statistical analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC, USA).
perforated appendicitis on abdominal CT scan. Of those found perforated in the OR, 17 (20%) were elderly (65 years or older), despite the elderly comprising only 8.4% (59/700) of our total cohort. The mean age of analyzed patients was 35.8 years (SD ¼ 15.2), and no correlation was found between perforation status and sex or ethnicity. Mean time to operation was 6.7 h (SD ¼ 4) and median time from antibiotic administration to appendectomy was 1.5 h (IQR: 0.2e3.6). Mean operative time was 56 min (SD ¼ 23), and all subjects underwent appendectomy within 27 h of presentation. Sample characteristics and demographic information can be found in Table 1. No relationship was found between time from admission to OR and in-hospital perforation. The timing of antibiotic administration also demonstrated no correlation to perforation status. It is important to note that transfer patients accounted for 26.6% (186/700) of the study cohort, yet comprised 37.6% (32/85) of all perforations. Baseline demographics were similar between groups, with a mean age of 36.5 years (SD ¼ 15.9) compared to 35.6 years (SD ¼ 14.8) in local patients. Transfer patients had a significantly increased rate of perforations (17.2% vs. 10.3%, p < 0.05), despite similar duration of symptoms at 26.4 h (SD ¼ 21) versus 26.1 h (SD ¼ 22.8) for the non-transfer group. The median time from antibiotic administration to OR was also similar at 1.1 h (IQR: 0.2e3.6) for the transfer group versus 1.6 h (IQR: 0.2e3.6). Surprisingly, time from admission to OR was shorter for the transferred patients at 5.8 h (SD ¼ 3.7) versus 7.1 h (SD ¼ 3.3) for the nontransfer cohort, which was significantly different (p < 0.01). Twenty patients waited longer than 18 h from admission to OR, of whom 7 (35%) were men and 13 (65%) women, and 2 (10%) were transfer patients. In 13 (65%) patients the CT scan findings were equivocal; those patients were given the option of a trial of antibiotic treatment versus surgery, and opted for the former. They were observed with IV antibiotics and upon showing no improvement of symptoms, subsequently received laparoscopic appendectomy. Of the remaining patients, two (10%) were women who presented with symptoms on postpartum day 1 after delivery, 1 (5%) has a history of kidney transplant with multiple comorbidities, 1 (5%) has a history of ulcerative colitis, 1 (5%) has a history of irritable bowel syndrome, all of which appeared to have confounded the initial diagnosis and caused untoward delay. Only 3 (15%) patients had no clear reason for a delay greater than 18 h until surgery. After adjusting for all covariates of interest, age >46 years, duration of symptoms, WBC, and neutrophil count were found to be significantly associated with in-hospital perforation (Table 2). Patients >46 years old were 4.5 times more likely to have perforation compared to those under 46 years of age (aOR ¼ 4.54, 95% CI ¼ 2.04e10.06, p < 0.001). Within this subset of patients aged 46 years or older, the risk of perforation appears to increase with age, with those 65 or older experiencing the highest rate of perforation (Table 3). In addition, a final diagnosis of perforation was associated with an increased average length of hospital stay across all age groups, with those 65 or older experiencing the longest hospital stays (Tables 3 and 4). Patients in whom the duration of symptoms was greater than 24 h were approximately two times more likely to have perforation compared to those patients in whom the duration of symptoms was less than or equal to 24 h (aOR ¼ 2.23, 95% CI ¼ 1.33e3.72, p < 0.001).
Results
Discussion
Of the 1106 charts reviewed, a total of 732 records met our inclusion criteria. Only 700 of those records were complete within the electronic medical record and available for analysis. Of these 700 patients, 85 (12%) were found to have a perforated appendix intraoperatively following a preoperative diagnosis of non-
Our goal was to evaluate what factors might contribute to the development of perforation in patients presenting with nonperforated acute appendicitis as confirmed by preoperative CT scan, with a specific focus on time from admission to OR and timing of antibiotic administration. Of note, our study demonstrated a
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
3
Table 1 Sample characteristics and demographics.
Sex, n (%) Female Male Age categories, n (%) 18e22 23e30 31e45 >46 Admission to OR- Time Categories*, n (%) 1st Quartile (<251) 2nd Quartile(251 - < 336) 3rd Quartile (336 - < 506) 4th Quartile (506) Antibiotics to OR- Time categories, n (%) No antibiotics 1st Quartile (>0 - <11) 2nd Quartile (11 - < 86) 3rd Quartile (86 - < 210) 4th Quartile (210) Duration of symptom till arrival, n (%) 24 h >24 h WBC (K/uL) Neutrophil (%seg)
No Perforation detected (n ¼ 615)
Perforation Detected (n ¼ 85)
355(57.72) 260(42.28)
47(55.29) 38(44.71)
133(21.63) 167(27.15) 179(29.11) 136(22.11)
9(10.59) 14(16.47) 23(27.06) 39(45.88)
158(25.69) 134(21.79) 184(29.92) 139(22.6)
25(29.41) 15(17.65) 28(32.94) 17(20)
48(7.8) 124(20.16) 142(23.09) 146(23.74) 155(25.2)
4(4.71) 18(21.18) 20(23.53) 19(22.35) 24(28.24)
462(75.12) 153(24.88) 14.3(12e17) 80(73e86)
53(62.35) 32(37.65) 15.8(13e18) 84.5(79e89)
P-value 0.67
<0.001
0.64
0.50
0.01 <0.003 <0.001
WBC and neutrophil variables are presented as median (IQR). Interval values for ‘Antibiotics to OR’ and ‘Admission to OR’ variables are given in minutes. Statistical tests used for group comparisons are discussed in the methods section.
significant correlation between perforation status and the age of the patient, as patients older than 46 had over a 4-fold increase in intra-operative diagnosis of perforated appendicitis. Of those patients >46 years old, the elderly experienced the highest rates of perforation and had a longer average length of stay. Symptom duration was also shown to be positively associated with the development of in-hospital perforation, as would be expected. Administration of antibiotics, time between antibiotic administration and admission to the OR, and time between hospital admission and appendectomy were not correlated to an outcome of inhospital perforation in our study population. These results suggest a benefit to expedient surgery in patients over the age of 46, with proportionately greater benefit as patient
age increases. On the other hand, the lack of correlation between in-hospital perforation and time from admission to OR suggests that delaying operations may not increase the risk of perforation in low-risk patients. In the literature focusing on risk factors for perforation or the development of complicated appendicitis, most authors focus on the effect of delaying surgery on perforation rates. Our finding of no significant correlation between time from hospital admission to OR and perforation status is largely supported by the available literature.11e15 Indeed, a recent meta-analysis on the topic found no increase in complicated appendicitis, postoperative surgical-site infection, or morbidity with delays up to 24hrs across 45 studies and 152,314 patients.16 In contrast, a retrospective cohort study by
Table 2 Unadjusted and Adjusted odds ratios for select variables.
Age categories 18-22 (reference) 23e30 31e45 >46 Antibiotics to OR 0 (reference) 1Q (>0 - < 11) 2Q (11 - < 86) 3Q (86 - < 210) 4Q (210) Admission to OR 1Q (reference) 2Q (251 - < 336) 3Q (336e506) 4Q (506) Duration of symptom 24 h (reference) >24 h WBC Neutrophil
OR
95% CI
p value
AOR
95% CI
p value
e 1.23 1.89 4.23
e (0.52e2.95) (0.85e4.23) (1.97e9.09)
e 0.62 0.11 <0.001
e 1.34 1.91 4.54
e (1.33e3.72) 0.55e3.25) (2.04e10.06)
e 0.52 0.13 <0.001
e 1.74 1.69 1.56 1.85
e (0.56e5.41) (0.55e5.19) (0.5e4.81) (0.61e5.62)
e 0.33 0.35 0.43 0.27
e 2.07 1.77 1.68 2.25
e (0.64e6.74) (0.55e5.72) (0.52e5.39) (0.69e7.37)
e 0.23 0.34 0.38 0.18
e 0.7 0.96 0.77
e (0.35e1.39) (0.53e1.71) (0.4e1.49)
e 0.31 0.89 0.44
e 0.76 0.97 0.7
e (0.37e1.56) (0.50e1.85) (0.32e1.55)
e 0.45 0.91 0.38
1.823 1.07 1.04
e (1.13e2.93) (1.02e1.13) (1.01e1.07)
e 0.01 <0.001 <0.001
e 2.23 1.08 1.03
e (1.33e3.72) (1.02e1.15) (1.04e1.06)
<0.001 0.01 0.02
Q: Quartile; OR: Unadjusted odds ratio; AOR: Adjusted odds ratio; CI: 95% Confidence Interval. Interval values for ‘Antibiotics to OR’ and ‘Admission to OR’ variables are given in minutes.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
4
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
Table 3 In-depth evaluation of 4th quartile age ranges.
Perforated (n) Non-Perforated (n) % Perforated Hospital Stay, Perf (days) Hospital Stay, Non-perf (days)
1st-3rd Quartile (n ¼ 515)
4th Quartile (n ¼ 175)
18e45
46e55
56e64
65þ
46 479 8.6% 3 (2e4) 1 (0e1)
11 71 13.4% 2 (2e3) 1 (1e2)
11 40 21.6% 3 (2e5) 1 (1e2)
17 25 40.5% 4 (3e6) 2 (1e2)
Hospital stay values are reported as median (IQR).
Ditillo et al.17 demonstrated a significantly increased risk of perforation when appendectomy is delayed, although the authors acknowledge that the duration of symptoms prior to admission have a more profound effect on perforation rates than in-hospital delays. Busch et al.18 also demonstrated increased perforation rates with appendectomy delay, but the findings were such that only delays >12 h were associated with a significantly higher frequency of perforated appendicitis (29.7 vs. 22.7%; P ¼ 0.010) whereas a delay of 6 or 9 h was not. Our average time from admission to OR was 6.7 h and only 7% of our cohort had a time to OR greater than 12 h (as opposed to their 19%), which suggests that our null findings are in fact consistent with those of Busch and colleagues, although the difference in patient sampling makes direct comparison less conclusive. There is additional literature on this topic in children which also suggest that delaying appendectomy within 24 h admission has no effect on perforation rates.13,19e21 However, these findings contrast with the results of Bonadio and colleagues.22 Their study showed delayed appendectomy significantly increased the risk of perforation in children who present with CT-documented uncomplicated appendicitis, although such selection criteria is prone to bias as CT is often avoided in children if ultrasound is diagnostic. Our finding that duration of symptoms and age correlate to perforation status agrees with much of the available literature,11,14,17,18 although the exact cutoff age for categorical age groupings varied between studies. The study by Busch et al.18 did not analyze duration of symptoms as an independent variable in their analysis, despite finding age to be an independent risk factor for perforation. It must be noted that some of these studies, such as those by Busch et al. and Kearner et al., are conducted outside of the United States, and therefore do not include CT-scan as part of the regular diagnostic process to rule out perforation in most patients. While their findings support age as an associated risk factor for a finding of perforation at the time of appendectomy, they do not strictly support age as a risk factor for perforation following admission to the hospital and a negative perforation status on CT scan, which is what we evaluated. Why duration of symptoms is consistently correlated to increased perforation but time from admission to appendectomy is not is likely the result of the small role of hospital delay in the total symptomatic interval. As mentioned previously, Ditillo and
colleagues17 did in fact find in-hospital delays to be an independent risk factor for perforation, but their findings showed that the duration of symptoms had the strongest predictive value when compared to the total symptomatic interval and time from admission to OR. It is also possible that their research, which stratified patients into 4 separate categories based on severity of disease, provided enough sensitivity for them to see an effect from in-hospital delays that other studies, including our own, missed. The results of our age analysis warrants further discussion, as the final age quartile has a much wider age range than the other quartiles, spanning from 46 to 88 years. While it is possible that elderly patients could have entirely accounted for the increased perforation rates and significant findings in this quartile, the median age for this category was 56 (IQR: 50e64) with 56% (22/39) of diagnosed perforations occurring in those younger than 65. So while the elderly did indeed contribute disproportionately to the amount of perforations in this quartile, we conclude that our findings are still relevant to older patients who are not yet elderly. A more detailed view of perforation status by age can be seen in Table 4. This finding that the elderly are more likely to experience perforation is supported in the available literature,23,24 but the reasons for this increased morbidity are not well established. It is known that the elderly often present with less obvious physical signs for many conditions, so it may be the case that the elderly patients in our cohort presented with a longer duration of symptoms. Physical diagnosis can also be more challenging in this population. However, our results show an average duration of symptoms of 21.6 h (SD ¼ 5.9) for the elderly, which is less than the 26.2 h average for our entire cohort. The mean time from admission to operation in the elderly was also slightly lower than the study average. Taken as a whole, these findings suggest a significant benefit to expedient appendectomy in older patients with appendicitis, and the corollary e potential harm from what might seem reasonable delays till surgery. As the US population ages, surgeons performing emergency general surgery will likely need to factor age in the decision on how best to mobilize resources for expedient appendectomy. Given the trend towards non-surgical management of acute appendicitis, we had hypothesized that antibiotic administration or the timing of such administration would be associated with perforation status. This was not seen to be the case in our study. It is
Table 4 Perforation status and length of hospital stay by age decile.
Age Range (years) Perforated (n) Non-perforated (n) % Perforated Hospital Stay, (Perforated, days) Hospital Stay, (Non-perforated, days)
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
18e20 2 68 2.9% 3.5 1 (1e1)
20e22 6 64 9.4% 3 1 (0e1)
23e25 5 65 7.7% 2 1 (0e1)
25e28 3 67 4.5% 5 1 (0e1)
28e32 11 59 18.6% 2 (1.5e2.5) 1 (1e2)
32e36 6 64 9.4% 3.5 1 (1e1)
36e42 6 64 9.4% 3 1 (1e2)
42e49 11 59 18.6% 3 (2.5e3) 1 (1e2)
49e59 9 61 14.8% 2 1 (1e2)
59e88 26 44 59.1% 4 (3e6) 1 (1e2)
Hospital stay values are reported as median (IQR). IQR values are only given for subsets with n > 10 as such analysis has little value in small datasets. n ¼ 70 for each decile.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
possible that antibiotic administration at our institution occurred within such a short time frame that no effect could be seen, as the mean time from antibiotic administration to OR was 2hr 30min and the median was 1hr 29min. While there is extensive research on the efficacy of antibiotic treatment relative to surgical management,6e10,25e27 there is limited research available on efficacy of antibiotics in preventing perforation prior to appendectomy. Studies on the efficacy of nonsurgical treatment found the antibiotic treatment failure rate within 24e48hrs to be 5%e11.9%, but none specifically explored perforation outcomes.28e30 Transfer patients had a shorter mean time from admission to OR in this study population. It is possible, and perhaps expected, that the initial workup by the transferring institution allowed for a more expedient processing by the receiving ER physicians and surgeons which resulted in more expedient operation. Despite this, transfer patients had a significantly higher rate of perforation, although all other analyzed variables were found to be similar between transfer and non-transfer patients. A possible explanation is that transfer patients had a longer overall duration of symptoms from time of onset to operation, given the additional time required to be transported to our institution which may have outweighed the benefit of a shorter time to operation after the patient arrived. We were not able to measure this variable precisely due to lack of adequate detail in the documentation recorded from outside institutions. It is also possible that patients from more rural communities had a higher rate of comorbid illness that was not accounted for by the variables measured in this study. Given its construction, we must label our study as having a low, IIIb evidence level based on the ELIS framework and Oxford Center for evidence-based medicine.31 Our study was a retrospective, casecontrol study. This study type increases the potential for a type I statistical error, and our positive results would likely require confirmation by more rigorous means, such as a prospective trial. In addition, there were some limitations regarding the collection and processing of our data. Information regarding the duration of symptoms was subject to bias, as values were collected from the emergency department physician note, and the values were often stratified to multiples of six. Duration of symptoms was also subject to bias in patient reporting. Our study was conducted at a single institution and is subject to the sampling biases inherent in recruiting subjects from a single location, although many of the related studies in the literature took a similar, single-institution approach. Our inclusion of transfer patients could have also impacted the broader applicability of our results given their higher rates of perforation and depressed average time from admission to OR when compared to non-transfer patients. These factors could result in an artificially decreased admission to OR time in our perforated population. However, save for those two variables, transfer patients shared the same patient characteristics as their non-transfer counterparts. Even with a significant difference between transfer and non-transfer time to OR, the addition of transfer patients only drops the mean time to OR from 7.3 h (SD ¼ 4.3) to 6.7 h (SD ¼ 4.3) for all perforated patients in this study. The similarities between transfer and non-transfer patients, in addition to the limited effect of the difference between them, lead us to believe that our findings for the entire cohort are sound, although we acknowledge transfers as a possible confounding factor. Regarding our diagnostic criteria, the diagnosis of non-perforated appendicitis on CT may not be 100% reliable, as we are assuming that the radiological diagnosis was uniformly correct. While CT scan has been shown to have great specificity in the diagnosis of perforated appendicitis, the sensitivity is less clear, although many studies report sensitivity >90% when using a combined criterion for perforation diagnosis.32e34 Finally, our ultimate diagnosis of perforation was taken from the attending surgeon's operative note,
5
not from the pathology report as has been reported in several studies in the literature. This could lead to missed cases of perforation and some inconsistency when comparing our results to similar studies that utilized the pathology report. Conclusions In this cohort of adults presenting with acute uncomplicated appendicitis, neither time to operation nor time to antibiotic administration were associated with in-hospital perforation. However, older patient age was associated with increased risk of inhospital perforation. Given that older patients also suffer increased consequences from perforation, there may be potential benefit from a “fast-track” process for elderly patients with appendicitis. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest The authors have no conflicts of interest to disclose. Acknowledgement The authors would like to acknowledge Rohali Keesari, MPH, PharmD for her contributions to the data analysis and the TTUHSC Clinical Research Institute. References 1. Williams GR. Presidential Address: a history of appendicitis. With anecdotes illustrating its importance. Ann Surg. 1983;197(5):495. 2. Margenthaler JA, Longo WE, Virgo KS, et al. Risk factors for adverse outcomes after the surgical treatment of appendicitis in adults. Ann Surg. 2003;238(1): 59e66. 3. Bhangu A, Søreide K, Di Saverio S, et al. Acute appendicitis: modern understanding of pathogenesis, diagnosis, and management. Lancet. 2015;386(10000):1278e1287. 4. Blomqvist PG, Andersson RE, Granath F, et al. Mortality after appendectomy in Sweden, 1987-1996. Ann Surg. 2001;233(4):455e460. 5. Vons C, Barry C, Maitre S, et al. Amoxicillin plus clavulanic acid versus appendicectomy for treatment of acute uncomplicated appendicitis: an openlabel, non-inferiority, randomised controlled trial. Lancet. 2011;377(9777): 1573. €rner U, Khorram-Manesh A, et al. Randomized clinical trial of 6. Hansson J, Ko antibiotic therapy versus appendicectomy as primary treatment of acute appendicitis in unselected patients. Br J Surg. 2009;96(5):473. €m L. Randomized controlled trial of appendicectomy versus 7. Eriksson S, Granstro antibiotic therapy for acute appendicitis. Br J Surg. 1995;82(2):166. 8. Styrud J, Eriksson S, Nilsson I, et al. Appendectomy versus antibiotic treatment in acute appendicitis. a prospective multicenter randomized controlled trial. World J Surg. 2006;30(6):1033. 9. Turhan AN, Kapan S, Kütükçü E, et al. Comparison of operative and non operative management of acute appendicitis. Ulus Travma Acil Cerrahi Derg. 2009;15(5):459e462. 10. Salminen P, Paajanen H, Rautio T, et al. Antibiotic therapy vs appendectomy for treatment of uncomplicated acute appendicitis: the APPAC randomized clinical trial. JAMA. 2015;313(23):2340. 11. Teixeira PG, Sivrikoz E, Inaba K, et al. Appendectomy timing: waiting until the next morning increases the risk of surgical site infections. Ann Surg. 2012;256(3):538e543. 12. Kim SH, Park SJ, Park YY, Choi SI. Delayed appendectomy is safe in patients with acute nonperforated appendicitis. Int Surg. 2015;100(6):1004e1010. 13. Kim M, Kim SJ, Cho HJ. Effect of surgical timing and outcomes for appendicitis severity. Ann Surg Treat Res. 2016;91(2):85e89. 14. Kearney D, Cahill RA, O'Brien E, et al. Influence of delays on perforation risk in adults with acute appendicitis. Dis Colon Rectum. 2008;51:1823e1827. 15. Kim JW, Shin DW, Kim DJ, et al. Effects of timing of appendectomy on the risks of perforation and postoperative complications of acute appendicitis. World J Surg. 2018;42:1295e1303. 16. van Dijk ST, van Dijk AH, Dijkgraaf MG, Boermeester MA. Meta-analysis of inhospital delay before surgery as a risk factor for complications in patients with acute appendicitis. Br J Surg. 2018;105:933e945.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
6
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
17. Ditillo MF, Dziura JD, Rabinovici R. Is it safe to delay appendectomy in adults with acute appendicitis? Ann Surg. 2006;244(5):656e660. 18. Busch M, Gutzwiller FS, Aellig S, et al. In-hospital delay increases the risk of perforation in adults with appendicitis. World J Surg. 2011;35:1626e1633. 19. Surana R, Quinn F, Puri P. Is it necessary to perform appendicectomy in the middle of the night in children? BMJ. 1993;306(6886):1168. 20. Cameron DB, Williams R, Geng Y, et al. Time to appendectomy for acute appendicitis: a systematic review. J Pediatr Surg. 2018;53:396e405. 21. Yardeni D, Hirschl RB, Drongowski RA, et al. Delayed versus immediate surgery in acute appendicitis: do we need to operate during the night? J Pediatr Surg. 2004;39(3):464e469. 22. Bonadio W, Brazg J, Telt N, et al. Impact of in-hospital timing to appendectomy on perforation rates in children with appendicitis. J Emerg Med. 2015;49(5): 597e604. 23. Horattas MC, Guyton DP, Wu D. A reappraisal of appendicitis in the elderly. Am J Surg. 1990;160(3):291e293. 24. Segev L, Keidar A, Schrier I, et al. Acute appendicitis in the elderly in the twenty-first century. J Gastrointest Surg. 2015 Apr;19(4):730e735. 25. Antibiotic therapy for acute appendicitis in adults. Fewer immediate complications than with surgery, but more subsequent failures. Prescrire Int. 2014;23(150):158e160. 26. Sallinen V, Akl EA, You JJ, et al. Meta-analysis of antibiotics versus appendicectomy for non-perforated acute appendicitis. Br J Surg. 2016;103(6): 656e667. 27. Harnoss JC, Zelienka I, Probst P, et al. Antibiotics versus surgical therapy for
28.
29.
30.
31.
32. 33.
34.
uncomplicated appendicitis: systematic review and meta-analysis of controlled trials (PROSPERO 2015: CRD42015016882). Ann Surg. 2017;265(5): 889e900. Rollins KE, Varadhan KK, Neal KR, et al. Antibiotics versus appendicectomy for the treatment of uncomplicated acute appendicitis: an updated meta-analysis of randomised controlled trials. World J Surg. 2016;40:2305e2318. Di Saverio S, Sibilio A, Giorgini E, et al. The NOTA study (Non-Operative Treatment for Acute Appendicitis): prospective study on the efficacy and safety of antibiotics (amoxicillin and clavulanic acid) for treating patients with right lower quadrant abdominal pain and long-term follow-up of conservatively treated suspected appendicitis. Ann Surg. 2014;260(1):109e117. Varadhan KK, Neal KR, Lobo DN. Safety and efficacy of antibiotics compared with appendicectomy for treatment of uncomplicated acute appendicitis: meta-analysis of randomised controlled trials. BMJ. 2012;344:e2156. Sauaia A, Moore EE, Crebs JL, et al. Evidence level of individual studies: a proposed framework for surgical research. J Trauma Acute Care Surg. 2012;72: 1484e1490. Horrow MM, White DS, Horrow JC. Differentiation of perforated from nonperforated appendicitis at CT. Radiology. 2003;227(1):46e51. Leite NP, Pereira JM, Cunha R, et al. CT evaluation of appendicitis and its complications: imaging techniques and key diagnostic findings. AJR Am J Roentgenol. 2005;185(2):406e417. Kim HY, Park JH, Lee YJ, et al. Systematic Review and Meta-analysis of CT features for differentiating complicated and uncomplicated appendicitis. Radiology. 2018;287(1):104e115.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
Contents lists available at ScienceDirect
The American Journal of Surgery journal homepage: www.americanjournalofsurgery.com
In-hospital perforation risk in acute appendicitis: Age matters Keith A. Hanson, Daron Jacob, Adel Alhaj Saleh, Sharmila Dissanaike* Department of Surgery, Texas Tech University Health Sciences Center, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 March 2019 Received in revised form 20 May 2019 Accepted 28 May 2019
Background: Controversy exists regarding how quickly an adult with appendicitis requires surgery to prevent perforation, and recent literature on antibiotic use as definitive treatment has complicated this question further. Since perforation is associated with worse outcomes, particularly in the elderly, efforts to prevent this complication are warranted. We studied risk factors for in-hospital perforation in patients diagnosed by admission CT with non-perforated acute appendicitis. Methods: We evaluated baseline demographics, symptom duration, and time from admission to antibiotics and surgery. Outcome measure was perforation diagnosed intra-operatively by attending surgeon. Results: Of 700 patients, 84 (12%) sustained in-hospital perforation; time from admission to operation or antibiotics were not associated. Duration of symptoms >24 h (aOR ¼ 2.23, 95% CI ¼ 1.33e3.72, p < 0.001) increased perforation risk. Patient age over 46 years (aOR ¼ 4.54, 95% CI ¼ 2.04e10.06, p < 0.001) was also associated with higher risk that increased with increasing age. Conclusion: Time to operation and antibiotic timing were not associated with in-hospital perforation in a general adult population. However, these findings suggest a possible benefit to expedient surgery in older patients. © 2019 Elsevier Inc. All rights reserved.
Introduction Acute appendicitis is one of the most common surgical emergencies worldwide.1 Perforation of the appendix, which increases the risk of abscess formation, wound infection, and sepsis, is a major source of morbidity associated with the condition.2,3 Perforation also increases mortality, with the risk of death varying from double the risk in young, otherwise healthy individuals to eight times the risk in the elderly when compared to uncomplicated appendicitis.4 While appendectomy is generally accepted as the treatment of choice, controversy exists regarding how quickly an adult with appendicitis requires surgery to prevent perforation and post-operative complications. The recent literature on antibiotic use as definitive treatment for appendicitis has complicated this question further, since it is theoretically feasible that antibiotic administration prevents perforation in appendicitis.5e10 Our objective in this study was to determine which variables influenced in-hospital perforation in adults who initially presented to the emergency room with uncomplicated, acute appendicitis as confirmed by CT scan. We hypothesized that a longer time from
* Corresponding author. 3601 4th St, Lubbock, TX, 79424, USA. E-mail address: [email protected] (S. Dissanaike).
admission to surgery would be associated with an increased incidence of perforation in patients with an initial diagnosis of nonperforated appendicitis. Additionally, we wanted to test whether timing of antibiotic administration affected the occurrence of inhospital perforation in these patients. Methods A retrospective chart review was conducted at an academic teaching hospital of patients who underwent appendectomy with a presumptive diagnosis of acute, non-perforated appendicitis, between January 2012 and December 2017. After IRB approval was obtained, ICD-9 code 540.9 (acute appendicitis) was used to procure a list of patient records for the given time frame. Patients who were found to have a perforated appendix on their initial CT scan were excluded from the study. Pregnant women and children were also excluded from the study, and subjects were required to have been between 18 and 89 years of age at the time of the operation. Transfer patients were included if adequate documentation was available from the outside facility. The majority of transfers were from local urgent care centers, freestanding emergency rooms and rural hospitals within a 100 mile radius. Data collected included perforation status, segmented neutrophil count, WBC count, duration of symptoms, sex, ethnicity, timing from antibiotic
https://doi.org/10.1016/j.amjsurg.2019.05.015 0002-9610/© 2019 Elsevier Inc. All rights reserved.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
2
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
administration to OR, and timing from admission to OR. Regular treatment protocol for appendicitis at this institution is for surgery to be consulted immediately on any patient with suspected appendicitis. Most patients receive a CT scan to confirm the diagnosis by the ER physician prior to consult. Dedicated in-house emergency general surgeons are available 24/7/365, therefore there is no delay in treatment due to conflict with elective surgery schedules; however, there may be delay due to more emergent cases. The practice of the surgical group at our institution is to offer laparoscopic appendectomy to all patients diagnosed with acute appendicitis. In order to obtain true informed consent, patients are also informed of the option of non-operative management with antibiotics, however the standard surgeon recommendation is for operative treatment. Occasionally a trial of non-operative management may be offered at attending surgeon discretion after extensive discussion with the patient in cases of true clinical equipoise as to the diagnosis; this is the exception, not the norm. Abdominal CT scans with or without IV contrast were reported by experienced, board-certified radiologists, and these reports were used as the definitive preoperative diagnosis (perforated vs. non-perforated appendicitis). The radiological criteria for perforation on CT was defined as the criteria for appendicitis in addition to one or more of the following: intra-abdominal abscess, extraluminal air, or extra-luminal appendicolith. If perforation was explicitly stated or reported that it could not be ruled out on the radiologist report of the CT, that patient was excluded from the study. If the radiologist's report made no mention of the possibility of perforation, then the patient was considered non-perforated on admission, and included as a subject in the study. The determination of whether a subject experienced perforation between hospital admission and OR was based on OR findings described in the operative notes from the board-certified attending surgeons performing or supervising the appendectomy. Admission time was based on the emergency department nursing note, and all data were collected using electronic medical records. Differences between the perforated and unperforated group were determined using the Student t-test and Wilcoxon rank sum test as appropriate for white blood cell counts and segmented neutrophil counts. The relationship between WBC, neutrophil counts and perforation status were too weak to justify further analysis by value categories with establishment of cutoff values. Chi-square or Fishers exact test were used for the variables race, sex, age, time from admission to OR, time from antibiotic administration to OR, and duration of symptoms. These variables were divided into quartiles and analyzed by category both to maintain as much statistical rigor as possible and to better establish cutoffs that would be more clinically relevant. Reported differences between transfer and non-transfer patients were determined using the Student t-test and Wilcoxon rank sum test as appropriate for all variables. Unadjusted logistic regression analysis was conducted to determine the association between individual predictors and inhospital perforation. Based on the maximum likelihood estimates, only those variables that showed significant associations (p < 0.05) with the outcome were considered for the adjusted logistic regression model. All statistical analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC, USA).
perforated appendicitis on abdominal CT scan. Of those found perforated in the OR, 17 (20%) were elderly (65 years or older), despite the elderly comprising only 8.4% (59/700) of our total cohort. The mean age of analyzed patients was 35.8 years (SD ¼ 15.2), and no correlation was found between perforation status and sex or ethnicity. Mean time to operation was 6.7 h (SD ¼ 4) and median time from antibiotic administration to appendectomy was 1.5 h (IQR: 0.2e3.6). Mean operative time was 56 min (SD ¼ 23), and all subjects underwent appendectomy within 27 h of presentation. Sample characteristics and demographic information can be found in Table 1. No relationship was found between time from admission to OR and in-hospital perforation. The timing of antibiotic administration also demonstrated no correlation to perforation status. It is important to note that transfer patients accounted for 26.6% (186/700) of the study cohort, yet comprised 37.6% (32/85) of all perforations. Baseline demographics were similar between groups, with a mean age of 36.5 years (SD ¼ 15.9) compared to 35.6 years (SD ¼ 14.8) in local patients. Transfer patients had a significantly increased rate of perforations (17.2% vs. 10.3%, p < 0.05), despite similar duration of symptoms at 26.4 h (SD ¼ 21) versus 26.1 h (SD ¼ 22.8) for the non-transfer group. The median time from antibiotic administration to OR was also similar at 1.1 h (IQR: 0.2e3.6) for the transfer group versus 1.6 h (IQR: 0.2e3.6). Surprisingly, time from admission to OR was shorter for the transferred patients at 5.8 h (SD ¼ 3.7) versus 7.1 h (SD ¼ 3.3) for the nontransfer cohort, which was significantly different (p < 0.01). Twenty patients waited longer than 18 h from admission to OR, of whom 7 (35%) were men and 13 (65%) women, and 2 (10%) were transfer patients. In 13 (65%) patients the CT scan findings were equivocal; those patients were given the option of a trial of antibiotic treatment versus surgery, and opted for the former. They were observed with IV antibiotics and upon showing no improvement of symptoms, subsequently received laparoscopic appendectomy. Of the remaining patients, two (10%) were women who presented with symptoms on postpartum day 1 after delivery, 1 (5%) has a history of kidney transplant with multiple comorbidities, 1 (5%) has a history of ulcerative colitis, 1 (5%) has a history of irritable bowel syndrome, all of which appeared to have confounded the initial diagnosis and caused untoward delay. Only 3 (15%) patients had no clear reason for a delay greater than 18 h until surgery. After adjusting for all covariates of interest, age >46 years, duration of symptoms, WBC, and neutrophil count were found to be significantly associated with in-hospital perforation (Table 2). Patients >46 years old were 4.5 times more likely to have perforation compared to those under 46 years of age (aOR ¼ 4.54, 95% CI ¼ 2.04e10.06, p < 0.001). Within this subset of patients aged 46 years or older, the risk of perforation appears to increase with age, with those 65 or older experiencing the highest rate of perforation (Table 3). In addition, a final diagnosis of perforation was associated with an increased average length of hospital stay across all age groups, with those 65 or older experiencing the longest hospital stays (Tables 3 and 4). Patients in whom the duration of symptoms was greater than 24 h were approximately two times more likely to have perforation compared to those patients in whom the duration of symptoms was less than or equal to 24 h (aOR ¼ 2.23, 95% CI ¼ 1.33e3.72, p < 0.001).
Results
Discussion
Of the 1106 charts reviewed, a total of 732 records met our inclusion criteria. Only 700 of those records were complete within the electronic medical record and available for analysis. Of these 700 patients, 85 (12%) were found to have a perforated appendix intraoperatively following a preoperative diagnosis of non-
Our goal was to evaluate what factors might contribute to the development of perforation in patients presenting with nonperforated acute appendicitis as confirmed by preoperative CT scan, with a specific focus on time from admission to OR and timing of antibiotic administration. Of note, our study demonstrated a
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
3
Table 1 Sample characteristics and demographics.
Sex, n (%) Female Male Age categories, n (%) 18e22 23e30 31e45 >46 Admission to OR- Time Categories*, n (%) 1st Quartile (<251) 2nd Quartile(251 - < 336) 3rd Quartile (336 - < 506) 4th Quartile (506) Antibiotics to OR- Time categories, n (%) No antibiotics 1st Quartile (>0 - <11) 2nd Quartile (11 - < 86) 3rd Quartile (86 - < 210) 4th Quartile (210) Duration of symptom till arrival, n (%) 24 h >24 h WBC (K/uL) Neutrophil (%seg)
No Perforation detected (n ¼ 615)
Perforation Detected (n ¼ 85)
355(57.72) 260(42.28)
47(55.29) 38(44.71)
133(21.63) 167(27.15) 179(29.11) 136(22.11)
9(10.59) 14(16.47) 23(27.06) 39(45.88)
158(25.69) 134(21.79) 184(29.92) 139(22.6)
25(29.41) 15(17.65) 28(32.94) 17(20)
48(7.8) 124(20.16) 142(23.09) 146(23.74) 155(25.2)
4(4.71) 18(21.18) 20(23.53) 19(22.35) 24(28.24)
462(75.12) 153(24.88) 14.3(12e17) 80(73e86)
53(62.35) 32(37.65) 15.8(13e18) 84.5(79e89)
P-value 0.67
<0.001
0.64
0.50
0.01 <0.003 <0.001
WBC and neutrophil variables are presented as median (IQR). Interval values for ‘Antibiotics to OR’ and ‘Admission to OR’ variables are given in minutes. Statistical tests used for group comparisons are discussed in the methods section.
significant correlation between perforation status and the age of the patient, as patients older than 46 had over a 4-fold increase in intra-operative diagnosis of perforated appendicitis. Of those patients >46 years old, the elderly experienced the highest rates of perforation and had a longer average length of stay. Symptom duration was also shown to be positively associated with the development of in-hospital perforation, as would be expected. Administration of antibiotics, time between antibiotic administration and admission to the OR, and time between hospital admission and appendectomy were not correlated to an outcome of inhospital perforation in our study population. These results suggest a benefit to expedient surgery in patients over the age of 46, with proportionately greater benefit as patient
age increases. On the other hand, the lack of correlation between in-hospital perforation and time from admission to OR suggests that delaying operations may not increase the risk of perforation in low-risk patients. In the literature focusing on risk factors for perforation or the development of complicated appendicitis, most authors focus on the effect of delaying surgery on perforation rates. Our finding of no significant correlation between time from hospital admission to OR and perforation status is largely supported by the available literature.11e15 Indeed, a recent meta-analysis on the topic found no increase in complicated appendicitis, postoperative surgical-site infection, or morbidity with delays up to 24hrs across 45 studies and 152,314 patients.16 In contrast, a retrospective cohort study by
Table 2 Unadjusted and Adjusted odds ratios for select variables.
Age categories 18-22 (reference) 23e30 31e45 >46 Antibiotics to OR 0 (reference) 1Q (>0 - < 11) 2Q (11 - < 86) 3Q (86 - < 210) 4Q (210) Admission to OR 1Q (reference) 2Q (251 - < 336) 3Q (336e506) 4Q (506) Duration of symptom 24 h (reference) >24 h WBC Neutrophil
OR
95% CI
p value
AOR
95% CI
p value
e 1.23 1.89 4.23
e (0.52e2.95) (0.85e4.23) (1.97e9.09)
e 0.62 0.11 <0.001
e 1.34 1.91 4.54
e (1.33e3.72) 0.55e3.25) (2.04e10.06)
e 0.52 0.13 <0.001
e 1.74 1.69 1.56 1.85
e (0.56e5.41) (0.55e5.19) (0.5e4.81) (0.61e5.62)
e 0.33 0.35 0.43 0.27
e 2.07 1.77 1.68 2.25
e (0.64e6.74) (0.55e5.72) (0.52e5.39) (0.69e7.37)
e 0.23 0.34 0.38 0.18
e 0.7 0.96 0.77
e (0.35e1.39) (0.53e1.71) (0.4e1.49)
e 0.31 0.89 0.44
e 0.76 0.97 0.7
e (0.37e1.56) (0.50e1.85) (0.32e1.55)
e 0.45 0.91 0.38
1.823 1.07 1.04
e (1.13e2.93) (1.02e1.13) (1.01e1.07)
e 0.01 <0.001 <0.001
e 2.23 1.08 1.03
e (1.33e3.72) (1.02e1.15) (1.04e1.06)
<0.001 0.01 0.02
Q: Quartile; OR: Unadjusted odds ratio; AOR: Adjusted odds ratio; CI: 95% Confidence Interval. Interval values for ‘Antibiotics to OR’ and ‘Admission to OR’ variables are given in minutes.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
4
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
Table 3 In-depth evaluation of 4th quartile age ranges.
Perforated (n) Non-Perforated (n) % Perforated Hospital Stay, Perf (days) Hospital Stay, Non-perf (days)
1st-3rd Quartile (n ¼ 515)
4th Quartile (n ¼ 175)
18e45
46e55
56e64
65þ
46 479 8.6% 3 (2e4) 1 (0e1)
11 71 13.4% 2 (2e3) 1 (1e2)
11 40 21.6% 3 (2e5) 1 (1e2)
17 25 40.5% 4 (3e6) 2 (1e2)
Hospital stay values are reported as median (IQR).
Ditillo et al.17 demonstrated a significantly increased risk of perforation when appendectomy is delayed, although the authors acknowledge that the duration of symptoms prior to admission have a more profound effect on perforation rates than in-hospital delays. Busch et al.18 also demonstrated increased perforation rates with appendectomy delay, but the findings were such that only delays >12 h were associated with a significantly higher frequency of perforated appendicitis (29.7 vs. 22.7%; P ¼ 0.010) whereas a delay of 6 or 9 h was not. Our average time from admission to OR was 6.7 h and only 7% of our cohort had a time to OR greater than 12 h (as opposed to their 19%), which suggests that our null findings are in fact consistent with those of Busch and colleagues, although the difference in patient sampling makes direct comparison less conclusive. There is additional literature on this topic in children which also suggest that delaying appendectomy within 24 h admission has no effect on perforation rates.13,19e21 However, these findings contrast with the results of Bonadio and colleagues.22 Their study showed delayed appendectomy significantly increased the risk of perforation in children who present with CT-documented uncomplicated appendicitis, although such selection criteria is prone to bias as CT is often avoided in children if ultrasound is diagnostic. Our finding that duration of symptoms and age correlate to perforation status agrees with much of the available literature,11,14,17,18 although the exact cutoff age for categorical age groupings varied between studies. The study by Busch et al.18 did not analyze duration of symptoms as an independent variable in their analysis, despite finding age to be an independent risk factor for perforation. It must be noted that some of these studies, such as those by Busch et al. and Kearner et al., are conducted outside of the United States, and therefore do not include CT-scan as part of the regular diagnostic process to rule out perforation in most patients. While their findings support age as an associated risk factor for a finding of perforation at the time of appendectomy, they do not strictly support age as a risk factor for perforation following admission to the hospital and a negative perforation status on CT scan, which is what we evaluated. Why duration of symptoms is consistently correlated to increased perforation but time from admission to appendectomy is not is likely the result of the small role of hospital delay in the total symptomatic interval. As mentioned previously, Ditillo and
colleagues17 did in fact find in-hospital delays to be an independent risk factor for perforation, but their findings showed that the duration of symptoms had the strongest predictive value when compared to the total symptomatic interval and time from admission to OR. It is also possible that their research, which stratified patients into 4 separate categories based on severity of disease, provided enough sensitivity for them to see an effect from in-hospital delays that other studies, including our own, missed. The results of our age analysis warrants further discussion, as the final age quartile has a much wider age range than the other quartiles, spanning from 46 to 88 years. While it is possible that elderly patients could have entirely accounted for the increased perforation rates and significant findings in this quartile, the median age for this category was 56 (IQR: 50e64) with 56% (22/39) of diagnosed perforations occurring in those younger than 65. So while the elderly did indeed contribute disproportionately to the amount of perforations in this quartile, we conclude that our findings are still relevant to older patients who are not yet elderly. A more detailed view of perforation status by age can be seen in Table 4. This finding that the elderly are more likely to experience perforation is supported in the available literature,23,24 but the reasons for this increased morbidity are not well established. It is known that the elderly often present with less obvious physical signs for many conditions, so it may be the case that the elderly patients in our cohort presented with a longer duration of symptoms. Physical diagnosis can also be more challenging in this population. However, our results show an average duration of symptoms of 21.6 h (SD ¼ 5.9) for the elderly, which is less than the 26.2 h average for our entire cohort. The mean time from admission to operation in the elderly was also slightly lower than the study average. Taken as a whole, these findings suggest a significant benefit to expedient appendectomy in older patients with appendicitis, and the corollary e potential harm from what might seem reasonable delays till surgery. As the US population ages, surgeons performing emergency general surgery will likely need to factor age in the decision on how best to mobilize resources for expedient appendectomy. Given the trend towards non-surgical management of acute appendicitis, we had hypothesized that antibiotic administration or the timing of such administration would be associated with perforation status. This was not seen to be the case in our study. It is
Table 4 Perforation status and length of hospital stay by age decile.
Age Range (years) Perforated (n) Non-perforated (n) % Perforated Hospital Stay, (Perforated, days) Hospital Stay, (Non-perforated, days)
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
18e20 2 68 2.9% 3.5 1 (1e1)
20e22 6 64 9.4% 3 1 (0e1)
23e25 5 65 7.7% 2 1 (0e1)
25e28 3 67 4.5% 5 1 (0e1)
28e32 11 59 18.6% 2 (1.5e2.5) 1 (1e2)
32e36 6 64 9.4% 3.5 1 (1e1)
36e42 6 64 9.4% 3 1 (1e2)
42e49 11 59 18.6% 3 (2.5e3) 1 (1e2)
49e59 9 61 14.8% 2 1 (1e2)
59e88 26 44 59.1% 4 (3e6) 1 (1e2)
Hospital stay values are reported as median (IQR). IQR values are only given for subsets with n > 10 as such analysis has little value in small datasets. n ¼ 70 for each decile.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
possible that antibiotic administration at our institution occurred within such a short time frame that no effect could be seen, as the mean time from antibiotic administration to OR was 2hr 30min and the median was 1hr 29min. While there is extensive research on the efficacy of antibiotic treatment relative to surgical management,6e10,25e27 there is limited research available on efficacy of antibiotics in preventing perforation prior to appendectomy. Studies on the efficacy of nonsurgical treatment found the antibiotic treatment failure rate within 24e48hrs to be 5%e11.9%, but none specifically explored perforation outcomes.28e30 Transfer patients had a shorter mean time from admission to OR in this study population. It is possible, and perhaps expected, that the initial workup by the transferring institution allowed for a more expedient processing by the receiving ER physicians and surgeons which resulted in more expedient operation. Despite this, transfer patients had a significantly higher rate of perforation, although all other analyzed variables were found to be similar between transfer and non-transfer patients. A possible explanation is that transfer patients had a longer overall duration of symptoms from time of onset to operation, given the additional time required to be transported to our institution which may have outweighed the benefit of a shorter time to operation after the patient arrived. We were not able to measure this variable precisely due to lack of adequate detail in the documentation recorded from outside institutions. It is also possible that patients from more rural communities had a higher rate of comorbid illness that was not accounted for by the variables measured in this study. Given its construction, we must label our study as having a low, IIIb evidence level based on the ELIS framework and Oxford Center for evidence-based medicine.31 Our study was a retrospective, casecontrol study. This study type increases the potential for a type I statistical error, and our positive results would likely require confirmation by more rigorous means, such as a prospective trial. In addition, there were some limitations regarding the collection and processing of our data. Information regarding the duration of symptoms was subject to bias, as values were collected from the emergency department physician note, and the values were often stratified to multiples of six. Duration of symptoms was also subject to bias in patient reporting. Our study was conducted at a single institution and is subject to the sampling biases inherent in recruiting subjects from a single location, although many of the related studies in the literature took a similar, single-institution approach. Our inclusion of transfer patients could have also impacted the broader applicability of our results given their higher rates of perforation and depressed average time from admission to OR when compared to non-transfer patients. These factors could result in an artificially decreased admission to OR time in our perforated population. However, save for those two variables, transfer patients shared the same patient characteristics as their non-transfer counterparts. Even with a significant difference between transfer and non-transfer time to OR, the addition of transfer patients only drops the mean time to OR from 7.3 h (SD ¼ 4.3) to 6.7 h (SD ¼ 4.3) for all perforated patients in this study. The similarities between transfer and non-transfer patients, in addition to the limited effect of the difference between them, lead us to believe that our findings for the entire cohort are sound, although we acknowledge transfers as a possible confounding factor. Regarding our diagnostic criteria, the diagnosis of non-perforated appendicitis on CT may not be 100% reliable, as we are assuming that the radiological diagnosis was uniformly correct. While CT scan has been shown to have great specificity in the diagnosis of perforated appendicitis, the sensitivity is less clear, although many studies report sensitivity >90% when using a combined criterion for perforation diagnosis.32e34 Finally, our ultimate diagnosis of perforation was taken from the attending surgeon's operative note,
5
not from the pathology report as has been reported in several studies in the literature. This could lead to missed cases of perforation and some inconsistency when comparing our results to similar studies that utilized the pathology report. Conclusions In this cohort of adults presenting with acute uncomplicated appendicitis, neither time to operation nor time to antibiotic administration were associated with in-hospital perforation. However, older patient age was associated with increased risk of inhospital perforation. Given that older patients also suffer increased consequences from perforation, there may be potential benefit from a “fast-track” process for elderly patients with appendicitis. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest The authors have no conflicts of interest to disclose. Acknowledgement The authors would like to acknowledge Rohali Keesari, MPH, PharmD for her contributions to the data analysis and the TTUHSC Clinical Research Institute. References 1. Williams GR. Presidential Address: a history of appendicitis. With anecdotes illustrating its importance. Ann Surg. 1983;197(5):495. 2. Margenthaler JA, Longo WE, Virgo KS, et al. Risk factors for adverse outcomes after the surgical treatment of appendicitis in adults. Ann Surg. 2003;238(1): 59e66. 3. Bhangu A, Søreide K, Di Saverio S, et al. Acute appendicitis: modern understanding of pathogenesis, diagnosis, and management. Lancet. 2015;386(10000):1278e1287. 4. Blomqvist PG, Andersson RE, Granath F, et al. Mortality after appendectomy in Sweden, 1987-1996. Ann Surg. 2001;233(4):455e460. 5. Vons C, Barry C, Maitre S, et al. Amoxicillin plus clavulanic acid versus appendicectomy for treatment of acute uncomplicated appendicitis: an openlabel, non-inferiority, randomised controlled trial. Lancet. 2011;377(9777): 1573. €rner U, Khorram-Manesh A, et al. Randomized clinical trial of 6. Hansson J, Ko antibiotic therapy versus appendicectomy as primary treatment of acute appendicitis in unselected patients. Br J Surg. 2009;96(5):473. €m L. Randomized controlled trial of appendicectomy versus 7. Eriksson S, Granstro antibiotic therapy for acute appendicitis. Br J Surg. 1995;82(2):166. 8. Styrud J, Eriksson S, Nilsson I, et al. Appendectomy versus antibiotic treatment in acute appendicitis. a prospective multicenter randomized controlled trial. World J Surg. 2006;30(6):1033. 9. Turhan AN, Kapan S, Kütükçü E, et al. Comparison of operative and non operative management of acute appendicitis. Ulus Travma Acil Cerrahi Derg. 2009;15(5):459e462. 10. Salminen P, Paajanen H, Rautio T, et al. Antibiotic therapy vs appendectomy for treatment of uncomplicated acute appendicitis: the APPAC randomized clinical trial. JAMA. 2015;313(23):2340. 11. Teixeira PG, Sivrikoz E, Inaba K, et al. Appendectomy timing: waiting until the next morning increases the risk of surgical site infections. Ann Surg. 2012;256(3):538e543. 12. Kim SH, Park SJ, Park YY, Choi SI. Delayed appendectomy is safe in patients with acute nonperforated appendicitis. Int Surg. 2015;100(6):1004e1010. 13. Kim M, Kim SJ, Cho HJ. Effect of surgical timing and outcomes for appendicitis severity. Ann Surg Treat Res. 2016;91(2):85e89. 14. Kearney D, Cahill RA, O'Brien E, et al. Influence of delays on perforation risk in adults with acute appendicitis. Dis Colon Rectum. 2008;51:1823e1827. 15. Kim JW, Shin DW, Kim DJ, et al. Effects of timing of appendectomy on the risks of perforation and postoperative complications of acute appendicitis. World J Surg. 2018;42:1295e1303. 16. van Dijk ST, van Dijk AH, Dijkgraaf MG, Boermeester MA. Meta-analysis of inhospital delay before surgery as a risk factor for complications in patients with acute appendicitis. Br J Surg. 2018;105:933e945.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015
6
K.A. Hanson et al. / The American Journal of Surgery xxx (xxxx) xxx
17. Ditillo MF, Dziura JD, Rabinovici R. Is it safe to delay appendectomy in adults with acute appendicitis? Ann Surg. 2006;244(5):656e660. 18. Busch M, Gutzwiller FS, Aellig S, et al. In-hospital delay increases the risk of perforation in adults with appendicitis. World J Surg. 2011;35:1626e1633. 19. Surana R, Quinn F, Puri P. Is it necessary to perform appendicectomy in the middle of the night in children? BMJ. 1993;306(6886):1168. 20. Cameron DB, Williams R, Geng Y, et al. Time to appendectomy for acute appendicitis: a systematic review. J Pediatr Surg. 2018;53:396e405. 21. Yardeni D, Hirschl RB, Drongowski RA, et al. Delayed versus immediate surgery in acute appendicitis: do we need to operate during the night? J Pediatr Surg. 2004;39(3):464e469. 22. Bonadio W, Brazg J, Telt N, et al. Impact of in-hospital timing to appendectomy on perforation rates in children with appendicitis. J Emerg Med. 2015;49(5): 597e604. 23. Horattas MC, Guyton DP, Wu D. A reappraisal of appendicitis in the elderly. Am J Surg. 1990;160(3):291e293. 24. Segev L, Keidar A, Schrier I, et al. Acute appendicitis in the elderly in the twenty-first century. J Gastrointest Surg. 2015 Apr;19(4):730e735. 25. Antibiotic therapy for acute appendicitis in adults. Fewer immediate complications than with surgery, but more subsequent failures. Prescrire Int. 2014;23(150):158e160. 26. Sallinen V, Akl EA, You JJ, et al. Meta-analysis of antibiotics versus appendicectomy for non-perforated acute appendicitis. Br J Surg. 2016;103(6): 656e667. 27. Harnoss JC, Zelienka I, Probst P, et al. Antibiotics versus surgical therapy for
28.
29.
30.
31.
32. 33.
34.
uncomplicated appendicitis: systematic review and meta-analysis of controlled trials (PROSPERO 2015: CRD42015016882). Ann Surg. 2017;265(5): 889e900. Rollins KE, Varadhan KK, Neal KR, et al. Antibiotics versus appendicectomy for the treatment of uncomplicated acute appendicitis: an updated meta-analysis of randomised controlled trials. World J Surg. 2016;40:2305e2318. Di Saverio S, Sibilio A, Giorgini E, et al. The NOTA study (Non-Operative Treatment for Acute Appendicitis): prospective study on the efficacy and safety of antibiotics (amoxicillin and clavulanic acid) for treating patients with right lower quadrant abdominal pain and long-term follow-up of conservatively treated suspected appendicitis. Ann Surg. 2014;260(1):109e117. Varadhan KK, Neal KR, Lobo DN. Safety and efficacy of antibiotics compared with appendicectomy for treatment of uncomplicated acute appendicitis: meta-analysis of randomised controlled trials. BMJ. 2012;344:e2156. Sauaia A, Moore EE, Crebs JL, et al. Evidence level of individual studies: a proposed framework for surgical research. J Trauma Acute Care Surg. 2012;72: 1484e1490. Horrow MM, White DS, Horrow JC. Differentiation of perforated from nonperforated appendicitis at CT. Radiology. 2003;227(1):46e51. Leite NP, Pereira JM, Cunha R, et al. CT evaluation of appendicitis and its complications: imaging techniques and key diagnostic findings. AJR Am J Roentgenol. 2005;185(2):406e417. Kim HY, Park JH, Lee YJ, et al. Systematic Review and Meta-analysis of CT features for differentiating complicated and uncomplicated appendicitis. Radiology. 2018;287(1):104e115.
Please cite this article as: Hanson KA et al., In-hospital perforation risk in acute appendicitis: Age matters, The American Journal of Surgery, https://doi.org/10.1016/j.amjsurg.2019.05.015