Prophylactic surgical drain placement with irrigation reduces abscess formation in patients with severe, uncontained, perforated appendicitis

Abstract

Background

20–25% of perforated appendicitis cases are complicated by abscess formation. This study assesses whether prophylactic closed-suction surgical drain (SD) placement after irrigation can decrease postoperative abscess formation in patients with extensively contaminated, perforated appendicitis.

Methods

A multicenter retrospective review was performed on pediatric patients with uncontained perforated appendicitis from January 1, 2020 to August 1, 2023. Limited irrigation was performed. All SDs were oriented towards the pelvis. Data were analyzed in four groups: (G1) SD and irrigation, (G2) only SD, (G3) only irrigation, (G4) neither SD nor irrigation.

Results

One hundred and fifteen patients (44 in G1, 3 in G2, 21 in G3, 47 in G4) were included. The abscess rate was 5.1% (G1), 33.3% (G2), 33.3% (G3), and 37.0% (G4) (p = 0.001). No interventional radiology (IR) drains were placed in G1 and G2. SD duration was 4.1 ± 2.1 days in G1 and 2.7 ± 0.6 days in G2. IR drain duration was 14 ± 10.7 days in G3 and 8.3 ± 3.9 days in G4 (p < 0.001). There were no significant differences in total length of stay and antibiotic duration.

Conclusions

Irrigation with prophylactic SD placement may prevent postoperative abscesses in patients with severe, uncontained, perforated appendicitis, but prospective data are needed.

Introduction

Acute appendicitis is the most common emergency surgical condition in pediatric patients, contributing to an estimated 70,000 hospital admissions annually. Perforation occurs in about 30% of these cases [1]. Notably, 20–25% of perforated appendicitis cases developed intraabdominal abscesses, which is associated with longer hospitalizations, longer duration of antibiotics, greater resource utilization, and further increases the risk of subsequent interventional procedures [2, 3]. Current management paradigms for intraabdominal abscesses secondary to perforated appendicitis center around source control, generally via interventional radiology (IR) drain placement and parenteral antibiotics. These approaches, though safe and effective, do carry up to a 5% complication rate and predispose the patient to another procedure, often requiring general anesthesia, depending on the patient’s age [4–6].

Previous studies evaluating prophylactic postoperative surgical drain (SD) placement in children with complicated appendicitis have failed to demonstrate a clear benefit. These studies, however, included open appendectomies and used non-suctioned drains (such as Penrose drains) or other gravity-dependent drainage systems [7–10]. This study aims to determine whether prophylactic surgical drain placement, with or without irrigation, can reduce the rate of abscess formation in cases of uncontained perforated acute appendicitis.

Methods

Inclusion/exclusion criteria

Following institutional review board approval, a retrospective chart review was performed on all cases of perforated appendicitis from January 1, 2020 to August 1, 2023 at two affiliated institutions. Three authors, blinded to outcomes, independently reviewed the operative notes using pre-established criteria to select cases to be included in the study [11]. The criteria for perforation included the presence of intraperitoneal appendicolith, fecal material, or a visible hole in the appendix. Uncontained perforation was identified by the presence of purulent or fecal material in more than one abdominal quadrant (Fig. 1). Exclusion criteria included cases where the perforation was contained to an abscess cavity, such as a retrocecal abscess, or if the spillage was limited to the right lower quadrant and pelvis only. Cases initially managed non-operatively, or those with upfront IR drainage followed by interval appendectomy were also excluded.

Fig. 1.

Representative intraoperative images demonstrating: A multiquadrant contamination, B intraoperative abscess cavity, C necrotic appendix with visible hole, D positioning of the surgical drain in the dependent portion of the pelvis

Operative techniques

All patients underwent standard laparoscopic appendectomy via a three-port technique. The appendix was ligated with either a stapler or an Endoloop (Ethicon, Johnson & Johnson, New Brunswick, NJ, USA). The decision to perform irrigation of the abdomen was determined by the operative surgeon; when performed, irrigation was focused on the areas of contamination and limited to less than 1000 ml of saline. All the patients in this study were operated on by six surgeons who worked at both institutions. When the contamination involved more than one quadrant, a closed-suction Blake drain was placed, oriented toward the pelvis (Fig. 1). Three of six surgeons in our group placed surgical drains when these criteria were met. The other three surgeons did not place surgical drains. Surgical drains were generally removed when output was clear, and the volume was less than 2 ml/kg.

Statistical methods

Demographic data were collected, including age at presentation, duration of symptoms, time from presentation to OR, antibiotic choice, duration of intravenous (IV) and oral antibiotics, and presenting white blood cell count (WBC). Outcomes of interest were postoperative abscess formation, IR drain placement, length of stay (LOS), total drain days, total duration of intravenous and oral antibiotics, and readmission rates. Patients were stratified into four groups for analysis: those with surgical drain (SD) and irrigation (Group 1), SD but no irrigation (Group 2), no SD but had irrigation (Group 3), no SD and no irrigation (Group 4). Data were expressed as mean with standard deviation and analyzed via ANOVA, with all statistical analyses performed using R statistical software (version 4.1.2), and statistical significance was determined at the 5% level.

Results

There were 115 total patients who met inclusion criteria from 346 charts reviewed. There were 44 patients in Group 1, 3 in Group 2, 21 in Group 3, and 47 in Group 4. Among the four groups, there were no statistically significant differences in sex, age at admission, presenting WBC, duration of symptoms, initial antibiotic choice, or choice of oral antibiotic on discharge as determined by the rounding surgeon (Table 1).

Table 1.

Demographic data

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47	p value
Sex (female), n (%)	16 (36.4)	3 (100)	10 (47.6)	21 (44.7)	0.20
Admission age (years) Mean (stdev)	9.1 (4.1)	13.7 (4.9)	8.8 (4.1)	10.3 (4.4)	0.12
BMI, Mean (stdev)	19.4 (4.5)	22.3 (0.9)	17.5 (2.6)	20.6 (5.0)	0.051
WBC at presentation (103/ml), Mean (stdev)	18.2 (6.5)	18.6 (7.8)	18.7 (6.7)	17.1 (5.7)	0.79
Location at presentation, n (%)					0.005
Treating hospital	35 (79.5)	1 (33.3)	13 (65)	22 (46.8)
Referring hospital	9 (20.5)	2 (66.7)	7 (35)	25 (53.2)
Symptom duration (days), Mean (stdev)	2.6 (1.5)	1.7 (0.6)	2.1 (1.0)	2.6 (1.4)	0.27
Initial imaging choice, n (%)
Ultrasound	16 (36.4)	1 (33.3)	13 (61.9)	32 (68.1)	0.011
CT	31 (70.5)	2 (66.7)	12 (57.1)	28 (59.6)	0.66
IV Abx choice, n (%)					0.59
Ceftriaxone/metronidazole	30 (68.2)	3 (100)	17 (81)	36 (76.6)
Piperacillin–Tazobactam	14 (31.8)	0 (0)	4 (19)	11 (23.4)
PO Abx choice–Augmentin n (%)	28 (93.3)	1 (100)	13 (92.9)	23 (88.5)	0.69

Intraoperatively, all patients in Groups 1–4 had multiquadrant contamination (p > 0.999). Group 1 had a higher finding of intraperitoneal fecaliths (p = 0.051) and intraoperative abscesses (p = 0.012). The visible hole in the appendix was similar across groups (p = 0.061) (Table 2).

Table 2.

Intraoperative markers of severity

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47	p value
Intraperitoneal fecalith, n (%)	16 (36.4)	0 (0)	5 (23.8)	6 (12.8)	0.051
Multiquadrant contamination, n (%)	44 (100)	3 (100)	21 (100)	47 (100)	> 0.999
Intraoperative abscess discovery, n (%)	24 (54.5)	1 (33.3)	6 (28.6)	11 (23.4)	0.012
Visible hole in the appendix, n (%)	44 (100)	2 (66.7)	20 (95.2)	45 (95.7)	0.061

The incidence of postoperative abscess formation was lowest in Group 1 at 6.8%, compared to 33.3% in Group 2, 33.3% in Group 3, and 38.3% in Group 4 (p = 0.001). The average surgical drain duration was 4.0 (± 2.0) days in Group 1 and 2.7 (± 0.6) days in Group 2. No IR drains were placed in Groups 1 and 2, whereas 9 IR drains were placed in 6 patients (28.6%) in Group 3, and 23 IR drains were placed in 16 patients (34%) in Group 4. For those needing an IR drain, the average drain duration was 14 (± 10.7) days in Group 3 and 8.4 (± 3.8) in Group 4. Across the entire cohort, including those that did not have a drain placed, there was no difference in average drain duration between groups (p = 0.66). However, when comparing patients who received a drain (either SD or IR) in all groups, the total duration of drains was significantly longer in Groups 3 and 4 (p < 0.001) (Table 3). Most IR drains were placed transabdominally, with other approaches including transrectal (6), transgluteal (4), and via the flank (2) (Table 4).

Table 3.

Abscess information and drain durations

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47	p value
Abscess, n (%)	3 (6.8)	1 (33.3)	7 (33.3)	18 (38.3)	0.001
Surgical drain duration (days), Mean (stdev)	4.0 (2.0)	2.7 (0.6)	–	–	0.19
Patients with IR drain, n (%)	0 (0)	0 (0)	6 (28.6)	16 (34.0)	< 0.001
IR drain duration (days), Mean (stdev)	N/A (N/A)	N/A (N/A)	14.0 (10.7)	8.4 (3.8)	0.28
Total drain duration (surgical + IR only) (days), Mean (stdev)	4.0 (2.0)	2.7 (0.6)	14.0 (10.7)	8.4 (3.8)	< 0.001
Average total drain duration across cohort (days), Mean (stdev)	4.0 (2.0)	2.7 (0.6)	4.0 (1.8)	2.9 (4.6)	0.66

Table 4.

IR drain descriptive data

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47
Number of patients with IR drains (n)	0	0	6	16
Total IR drains (n)	0	0	9	23
Transabdominal approach (n)	0	0	6	14
Transrectal approach (n)	0	0	0	6
Flank approach (n)	0	0	0	2
Transgluteal approach (n)	0	0	3	1

Group 1 had a shorter, but not statistically significant, LOS at 4.5 (± 1.9) days compared to 5.9 (± 2.9) days in Group 4 (p = 0.20). There were also no differences between the four groups in the percentage of patients who had an ileus and days until return of bowel function (p = 0.86 and 0.79, respectively). Groups 3 and 4 had higher rates of emergency room (ER) visits (p = 0.006) and readmission (p = 0.042) within the first 30 days postoperatively, although the combined LOS of initial hospitalization and readmission did not reach statistical significance (p = 0.19) (Table 5). There were no significant differences between groups in terms of ER visits and readmissions after the initial 30 days (p = 0.21 and 0.13, respectively).

Table 5.

Length of stay and readmission information

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47	p value
Ileus, n (%)	11 (25)	1 (33.3)	5 (25)	15 (31.9)	0.86
Ileus time (days), Mean (stdev)	4.4 (2.2)	3.0 (N/A)	4.8 (2.6)	4.7 (1.9)	0.79
Initial LOS (days), Mean (stdev)	4.5 (1.9)	4.7 (1.5)	4.8 (2.4)	5.9 (2.9)	0.20
Readmission LOS (days), Mean (stdev)	3.0 (N/A)	N/A (N/A)	12.3 (11.2)	2.2 (0.8)	0.05
Total LOS (combined initial and repeat) (days), Mean (stdev)	4.5 (1.9)	4.7 (1.5)	6.5 (6.4)	6.1 (3.1)	0.19
30-day ER visit, n (%)	1 (2.3)	1 (33.3)	6 (28.6)	9 (19.1)	0.006
30-day readmission, n (%)	1 (2.3)	1 (33.3)	4 (19.0)	5 (10.6)	0.042
1-month–1-year ER visit, n (%)	1 (2.3)	0 (0)	0 (0)	2 (4.3)	0.21
1-month–1-year readmission, n (%)	0 (0)	0 (0)	1 (4.8)	1 (2.1)	0.13

Group 1 had a mean IV antibiotic duration of 4.5 ± 1.8 days, whereas Group 2 was 5.3 ± 0.6 days, Group 3 was 6.5 ± 5.9 days, and Group 4 was 6.4 ± 4.1 days (p = 0.15). A greater percentage of patients in Groups 3 and 4 required additional courses of IV antibiotics, but this did not reach statistical significance (p = 0.14). The total antibiotic duration was not significantly different among the four groups (Table 6).

Table 6.

Antibiotic durations

	Group 1 Drain and irrigation n = 44	Group 2 Drain and no irrigation n = 3	Group 3 No drain and irrigation n = 21	Group 4 No drain and no irrigation n = 47	p value
Initial IV Abx duration (days), Mean (stdev)	4.5 (1.8)	5.3 (0.6)	6.5 (5.9)	6.4 (4.1)	0.15
Initial Total Abx duration (days), Mean (stdev)	7.8 (2.8)	8.7 (6.4)	8.0 (3.9)	8.9 (4.8)	0.96
Patients needing repeat Abx, n (%)	1 (2.3)	0 (0)	3 (14.3)	6 (12.8)	0.21
Readmission IV Abx duration (days), Mean (stdev)	3.0 (N/A)	N/A (N/A)	11.0 (9.6)	4.7 (6.1)	0.14
Readmission total Abx duration (days), Mean (stdev)	10 (N/A)	N/A (N/A)	18.3 (10.2)	12.5 (6.0)	0.63
Total Abx duration (days), Mean (stdev)	12.6 (4.4)	14.0 (6.9)	17.1 (13.9)	16.9 (9.7)	0.55

Discussion

In this retrospective chart review of 115 pediatric patients at two institutions with severe, uncontained, perforated appendicitis, those who received irrigation followed by strategic placement of prophylactic surgical drains had reduced rates of postoperative intraabdominal abscess formation, no need for IR drain placement, and fewer ER visits and readmissions compared to those who did not receive prophylactic surgical drain placement or surgical drain placement without irrigation. This study is the first to clearly demonstrate the benefits of selectively and strategically placing surgical drains after irrigation in patients with severe cases of perforated appendicitis. The process involved irrigating the abdomen and positioning the surgical drain at the lowest point in the pelvis, which facilitated the evacuation of diluted purulent fluid postoperatively, thereby reducing the bacterial burden and fluid load in the abdomen, factors we believe are critical to reducing the incidence of postoperative abscesses. Although 3/44 (6.8%) patients who received surgical drains with irrigation developed abscesses, the abscesses were small and did not require IR intervention.

These findings contrast with many published studies on the efficacy of prophylactic surgical drain placement in pediatric acute appendicitis. A significant limitation of prior studies is their broad patient selection, including all cases of perforated appendicitis rather than focusing mainly on the most severe cases [7, 8, 10, 12]. Anandalwar et al. documented in a NSQIP-P database study that the presence of a greater number of intraoperative findings—such as a visible hole, multiquadrant fibrinopurulent exudate, abscess, and extraluminal fecalith—are associated with poorer outcomes and greater resource utilization, including higher rates of surgical site infection, revisit rates, and length of stay [11]. Therefore, it stands to reason that this group of patients likely benefit most from prophylactic surgical drain placement. In our study, the high prevalence of multi-quadrant contamination, intraoperative abscesses, visible holes and intraperitoneal fecalith indicates a targeted selection of the most severe cases of perforated appendicitis. As a result, it is reasonable to infer that patients who received irrigation and prophylactic surgical drain (Group 1) experienced a reduction in abscess formation, no secondary procedures (i.e., IR) performed, and fewer readmissions compared to those who did not receive either surgical drain placement or irrigation (Groups 2–4).

Other limitations of prior studies involve drain type and location. Some prior studies placed the drain in the right lower quadrant or at the resection bed, and others utilized gravity-dependent drainage systems like Penrose drains [7, 8, 13, 14]. Studies employing closed-suction drains have demonstrated several benefits: shorter time to the resumption of a full diet, decreased postoperative LOS, and lower abscess rates [14]. Moreover, studies involving adult patients reported an abscess rate of approximately 6% in the prophylactic closed-suction drain group compared to 20% in the group without drains, and another study found a reduction in reintervention rates in the prophylactic closed-suction drain group [15, 16]. In our study, the surgical drain with irrigation group had an abscess rate of 6.8%, compared to 33.3–38.3% in other groups. This significant reduction suggests a synergistic effect of combined irrigation and closed-suction drainage in minimizing abscess formation. Interestingly, irrigation alone has not been shown to reduce abscess rates in perforated appendicitis—sometimes even increasing them [3, 17]—our study confirmed similarly high rates of abscess formation in Groups 3 at 33.3% when only irrigation was employed. However, we believe that irrigation helps decrease the viscosity of the purulent fluid and loosens the debris adherent to the mesentery, abdominal wall, or bowel loops, aiding removal via the dependently positioned pelvic surgical drain. Saline, as documented primarily in peritoneal dialysis literature, can irritate the peritoneum, and its chronic use can lead to increased adhesions and fibrosis [18, 19]. In limited exposure, however, the inflammatory effect of saline irrigation may temporarily increase peritoneal fluid volume, thereby further diluting residual bacteria and increasing the flow of effluent into the surgical drain.

Another significant finding of this study was the decreased utilization of IR to place drains in patients who received prophylactic surgical drains. In Group 1, there were only three total abscesses, all sub-centimeter abscesses that were not considered clinically significant. In one case in Group 2, the abscess was 2 cm and was deemed inaccessible for IR drainage. Conversely, in Groups 3 and 4, there were 9 and 23 drains placed in 6 and 21 patients, respectively. Furthermore, the duration of IR drains was significantly longer—ranging from 1 to 2 weeks—compared to 2–4 days for surgical drains.

From a patient comfort perspective, the surgical drains were inserted through an existing port site made during the appendectomy, whereas the IR drains were placed using several different approaches. In our cohort, 20 IR drains were placed transabdominally, 2 through the flank, 4 transgluteally, and 6 transrectally. Transgluteal drains, while effective, can be painful in pediatric patients and may restrict their mobility post-procedure [20]. Transrectal drains are prone to dislodgement during bowel movements and may cause bleeding from the rectal wall [21, 22]. Although generally well tolerated, these IR drains carry a risk of subsequent complications, reported in up to 5%, including bleeding, incomplete drainage requiring additional drains or procedures, and inadvertent catheterization of nearby structures [4, 5]. In addition, placing an IR drain in pediatric patients often requires general anesthesia, further increasing the risks. Furthermore, the process of diagnosing an abscess and draining it with IR can also increase the child’s exposure to ionizing radiation, unless ultrasound-guided methods are used [6]. Therefore, prophylactic surgical drain placement may minimize subsequent imaging studies, exposure to anesthesia, and secondary procedures.

There are several limitations of this study. First, its retrospective nature limits its generalizability. Second, drain and irrigation were used at the operative surgeon’s discretion, and not all surgeons in this cohort employed prophylactic surgical drains or irrigation as part of their management strategy for perforated appendicitis, resulting in variability among the groups. However, the similarity in baseline demographics and intraoperative factors across the groups allowed for a valid comparative analysis. Finally, the sample size was relatively small, especially in Group 2. The authors believe that the synergistic effect of prophylactic drainage and irrigation contributes significantly to reducing abscess formation, which explains the infrequency of prophylactic drain placement without irrigation. This small sample size does limit the conclusions that can be drawn from this study.

Conclusions

In appropriately selected pediatric patients with severe, uncontained, perforated appendicitis, the use of a prophylactic, closed-suction drain coupled with limited irrigation may minimize abscess formation. However, prospectively collected, ideally randomized, data are essential to further study if prophylactic surgical drains and irrigation have any benefits in this population.

Abbreviations

IR

Interventional radiology

IV

Intravenous

LOS

Length of stay

SD

Surgical drain

WBC

White blood cell

Data availability

No datasets were generated or analysed during the current study.