Nonoperative Management of Uncomplicated Acute Appendicitis: A Systematic Review and Meta‐Analysis of Randomized Clinical Trials Comparing Antibiotic Treatment and Appendectomy in Children and Adolescents

ABSTRACT

Background

This systematic review with meta‐analysis aims to evaluate the current evidence comparing antibiotic therapy with the surgical gold standard (appendectomy) for the treatment of uncomplicated acute appendicitis (UAA) in children and adolescents.

Methods

This systematic review and meta‐analysis followed PRISMA guidelines and was registered in PROSPERO, CRD420251011305. MEDLINE and CENTRAL were systematically searched up to March 2025. Randomized controlled trials (RCTs) comparing antibiotic therapy and appendectomy in patients aged 0–18 years with UAA were included. Summary data were extracted from published reports of eligible RCTs. Full text eligibility and data extraction were performed independently by two reviewers. Primary outcome was the complication‐free treatment success rate at 30 days. Secondary outcomes included recurrence and complicated appendicitis at surgery. Risk ratios (RR) with 95% CIs were calculated using a random effects model.

Results

Of 815 records screened, six studies met inclusion criteria and were included in the review. A clinically relevant difference emerged in terms of complication‐free treatment success between the two groups (1333 participants; RR 0.90, 95% CI 0.84–0.96; I ² 63%; and moderate certainty of evidence) favoring surgical management over antibiotic therapy among the five studies included in the meta‐analysis. Recurrence after successful NOM occurred in 6%–24% of patients at 1 year

Conclusion

Antibiotic therapy could represent a feasible treatment option for UAA in children and adolescents, although in terms of complication‐free treatment success, appendectomy remains the gold standard.

Trial Registration

CRD420251011305

This systematic review with meta‐analysis aims to evaluate the current evidence comparing antibiotic therapy with the surgical gold standard (appendectomy) for the treatment of uncomplicated acute appendicitis (UAA) in children and adolescents. Antibiotic therapy could represent a feasible treatment option for UAA in children and adolescents, although in terms of complication‐free treatment success, appendectomy remains the gold standard.

1. Introduction

Uncomplicated acute appendicitis (UAA) is the most common abdominal surgical emergency in children and adolescents, with an estimated global incidence of 100–150 cases per 100,000 person‐year in high‐income countries and lower, but rising, rates in low‐ and middle‐income countries [1, 2]. For more than a century, appendectomy has been the gold standard of treatment for acute appendicitis [3]. The advent of laparoscopic surgery and its proven advantages has reinforced surgery as the preferred approach, including pediatric populations [4].

Emerging evidence increasingly suggests that UAA is more benign than previously considered and that it does not inevitably progress to perforation if left untreated. Most patients with UAA can be treated successfully with antibiotics, supporting nonoperative management (NOM) [3, 4, 5]. NOM with antibiotics is gaining attention as an alternative to appendectomy also for selected pediatric patients with UAA. From randomized clinical trials (RCTs) and observational studies, it has been reported that NOM with antibiotics can lead to clinical resolution, with acceptable recurrence rates within shared decision‐making frameworks [6]. Systematic reviews with meta‐analysis have been conducted to confirm these results in this setting [7, 8, 9]. Nevertheless, some clinical questions remain still open and raise concerns due to the selection of heterogeneous parameters—for example, combination of pediatric and adult populations without separated results [10]—and the absence of composite, patient‐centered end points that better reflect the benefit of treatment [11]. In addition, recent high‐quality RCTs evidence, including the latest trial by St Peter et al., has not been consistently incorporated into previous syntheses. Therefore, an updated and methodologically rigorous synthesis focusing exclusively on pediatric RCTs is warranted.

The aim of this systematic review is to assess the available evidence on NOM with antibiotics for UAA in children and adolescents when compared to appendectomy, focusing on safety and efficacy, and to discuss the limitations of published RCTs, potentially limiting a more widespread adoption of the antibiotic therapy as the first treatment option.

2. Material and Methods

This systematic review has been conducted according to the recommendations of the Preferred Reporting Items for Systematic reviews and Meta‐Analyses (PRISMA) guidelines [12]. The protocol has been registered on the International Prospective Register of Systematic Reviews (PROSPERO), registration number CRD420251011305. We systematically searched MEDLINE and Cochrane Central Register of Controlled Trials (CENTRAL) for relevant studies. Reference lists of relevant studies were searched manually. Search strategy is reported in BOX 1. No restrictions were placed on publication status or language. Literature search was completed in March 2025.

BOX 1.

1.

Search strategy: (“uncomplicated acute appendicitis” OR “acute appendicitis” OR appendicitis) AND (“non‐operative treatment” OR “non‐operative management” OR NOM OR “conservative treatment” OR “antibiotic treatment”) AND (“laparoscopic appendectomy” OR “laparoscopic appendicectomy” OR appendectomy OR appendicectomy)

Search: (“uncomplicated acute appendicitis” OR “acute appendicitis” OR appendicitis) AND (“non‐operative treatment” OR “non‐operative management” OR NOM OR “conservative treatment” OR “antibiotic treatment”) AND (“laparoscopic appendectomy” OR “laparoscopic appendicectomy” OR appendectomy OR appendicectomy)

(“uncomplicated acute appendicitis” [All Fields] OR “acute appendicitis” [All Fields] OR (“appendical” [All Fields] OR “appendicitis” [MeSH Terms] OR “appendicitis” [All Fields])) AND (“non‐operative treatment” [All Fields] OR “non‐operative management” [All Fields] OR “NOM” [All Fields] OR “conservative treatment” [All Fields] OR “antibiotic treatment” [All Fields]) AND (“laparoscopic appendectomy” [All Fields] OR “laparoscopic appendicectomy” [All Fields] OR (“appendectomy” [MeSH Terms] OR “appendectomy” [All Fields] OR “appendectomies” [All Fields] OR “appendicectomies” [All Fields] OR “appendicectomy” [All Fields]) OR (“appendectomy” [MeSH Terms] OR “appendectomy” [All Fields] OR “appendectomies” [All Fields] OR “appendicectomies” [All Fields] OR “appendicectomy” [All Fields]))

37

We included RCTs evaluating children and adolescents (0–18) with UAA (defined as appendicitis without localized or diffuse peritonitis or abscess) aimed to compare NOM with antibiotics versus appendectomy (laparoscopic or open). NOM included intravenous and/or oral antibiotics, with appendectomy reserved for nonresponsiveness to antibiotic treatment or appendicitis recurrence. In the comparator group, appendectomy was performed within 24–48 h after randomization, while perioperative antibiotics were allowed but not as an initial NOM. Studies meeting the inclusion criteria must describe well‐defined treatment protocols and report at least two of our selected outcome measures.

The primary outcome was complication‐free treatment success rate at 30 days (intended as success of the initial treatment with uncomplicated course, i.e. no post intervention complications, adverse events, or non‐responsiveness to antibiotic treatment). Nonresponsiveness to antibiotic treatment in the NOM group at index admission and at 30 days was identified with need for an appendectomy.

Secondary outcomes based on 1‐year follow‐up were

• recurrent acute appendicitis with need for appendectomy after 30 days

• negative appendectomy rate (NAR)

• post‐intervention complications (intra‐abdominal abscess, postoperative peritonitis, surgical site infections (SSIs), incisional hernias, incisional pain or obstructive symptoms, adverse reactions to antibiotic therapy, and other general complications)

• postoperative complications, based on the number of patients who underwent surgery (patients who underwent appendectomy as primary treatment and those who underwent appendectomy after failure of antibiotic treatment during the hospitalization)

• length of hospital stay (LOS)

• health‐related quality of life (QoL)

• costs

Study identification, selection, quality assessment, and data abstraction were carried out independently by two reviewers (EA and BS). Two independent reviewers screened titles and abstracts to identify potentially eligible studies and confirmed eligibility by reading the full text publication. Any discrepancies were resolved by consensus by a third reviewer (MP). We extracted study identifier (first author and year of publication), essential study information (study period and location and study population), treatment arms and the number of enrolled subjects, baseline characteristics of study subjects (mean age, and sex), general characteristics of eligible studies (inclusion and exclusion criteria; NOM group: type of antibiotics, route, frequency, and total dose/day; surgery group: type of surgery, antibiotic treatment before and after surgery, and definition of the investigated primary and secondary outcomes), and treatment outcomes.

We assessed the risk of bias for the RCTs enrolled in the systematic review and meta‐analysis with the revised Cochrane risk‐of‐bias (RoB) tool for randomized trials (RoB 2) [13]. We considered the effect of assignment to the interventions at the baseline, regardless of whether the interventions were received as intended (“intention‐to‐treat effect”). We examined the certainty of evidence using the Grading of Recommendation, Assessment, Development and Evaluation (GRADE) for the outcomes included in the meta‐analyses [14].

We undertook meta‐analyses when treatments, participants, and the underlying clinical questions were similar enough for pooling, while we narratively described skewed data. The effect sizes were calculated by risk ratio (RR) for dichotomous variables, with 95% confidence intervals (95% CI). The point estimate of the RR value was considered statistically significant at P‐level of less than 0.05 if the 95% CI does not cross value 1.

Heterogeneity of the results across studies was assessed using the Higgins' I ² and χ ² tests. A p‐value of χ ² test less than 0.10 with an I ² value of greater than 50% was considered as indicative of substantial heterogeneity and greater than 75% as considerable heterogeneity [15]. A random effects model (DerSimonian and Laird method) was used to account variability across studies.

3. Results

The search identified 815 records, resulting in 208 unique records after the removal of duplicates and inclusion of two records through hand searching. After screening titles and abstracts, we included six RCTs aimed at comparing NOM with antibiotics versus appendectomy, as reported in Figure 1.

FIGURE 1.

PRISMA flowchart.

Three studies tested NOM superiority over appendectomy [16, 17, 18], two tested noninferiority [19, 20], and one was a feasibility study [21].

The included studies were conducted in seven high‐income countries (UK, USA, Australia, Singapore, Finland, Sweden, and Canada) and one low‐income country (Pakistan). Four studies were multicenter RCTs [16, 19, 20, 21]. The oldest trial started the recruitment process in 2012 [18], whereas the newest in 2016 [20].

In total, the six RCTs included 1485 participants: 769 participants were randomized to NOM while 716 to appendectomy.

All the included RCTs planned to include both children and adolescents, with a median age of 10 years (5–15.11 years). Table 1 shows patients' general characteristics.

TABLE 1.

Characteristics of included studies.

Study (first name, year)	n. Trial centers	Country	Mean age	Sample size	Intervention—NOM route, frequency, total dose/day	Control group	Follow‐up	Complication‐free treatment success up to 30 days definition	Funding
Hall 2021 [21]	3	United Kingdom	Median (range) NOM 10.3 (5.0–15.22) surgery 10.7 (6.4–13.6)	57	Minimum of 24‐h broad spectrum IV antibiotics (per local antimicrobial policy) If afebrile for 24 h—conversion to oral antibiotics (per local policy) Total 10 days antibiotics following randomization	Open or laparoscopic appendicectomy at the surgeon's discretion	6‐month	Safety and overall success of initial nonoperative treatment; Complications of disease and treatment (Measured during hospital stay)	UK National Institute for Health Research Health Technology Assessment program (Grant Number: 14/192/90)
Svensson 2015 [18]	1	Sweden	Median (range) NOM 11.2 (5.9–15.0) Surgery 11.1 (6.2–14.8)	51	IV meropenem 10 mg/kg per 3 per 24 h and metronidazole 20 mg/kg per 1 per 24 h for at least 48 h If clinically well—conversion to oral ciprofloxacin 20 mg/kg per 2 per 24 h and metronidazole 20 mg/kg per 1 per 24 h For another 8 days	Open or laparoscopic appendicectomy not stipulated in the trial protocol	12‐month 5‐year	Proportion of children in each Group achieving resolution of symptoms without significant complications	Supported by the Crown Princess Lovisa's foundation and the Hirisch Foundation Grants
St peter 2025 [20]	11	Canada, Finland, Sweden, Singapore, and USA	Median (IQR) NOM 10.6 (8.7–12.9) surgery 10.9 (8.7–13.5)	978	IV fluid treatment minimum of 12h of IV antibiotics Minimum of 12h of clear fluid only and regular clinical review Switch to oral antibiotics if clinical improvement shown Further 24 h if no improvement for discharge (protocol) Total course of 10 days of antibiotics (ciprofoloxacin and metronidazole or amoxicillin and clavulanic acid (intravenous and oral) following randomization (protocol)	Laparoscopic appendicectomy 18 h after randomization	6‐week 3‐month 12‐month	Treatment failure	Sweden—grant funding secured from Swedish Research Council and Stiftelsen Frimurare Barnhuset Canada—grant funding obtained from the Academic Medical Organisation of Southwestern Ontario
Adams 2024 [19]	2	Australia	Median (IQR): NOM 10.0 (8.0–11.0) Surgery 9.0 (8.0–11.0)	222	IV piperacillin with tazobactam eight‐hourly at 100 mg/kg/dose for at least 24 h—further 24 h of piperacillin with tazobactam if no improvement for discharge (protocol) (Replacement due to shortage) IV amoxicillin with clavulanic acid eight‐hourly at 25 mg/kg/dose After discharge—oral amoxicillin with clavulanic acid 22.5 mg/kg/dose If intolerance or allergy—oral ciprofloxacin 15 mg/kg/dose 2 times a day or oral metronidazole 10 mg/kg/dose 2 times a day (protocol)	Appendicectomy performed laparoscopically or open according to surgeon's standard practice	30 days 12 months	Treatment success, based on occurrence of (a) unplanned operations, (b) unnecessary operations, and (c) complications of the appendicitis itself in the NOM arm or due to the management of the appendicitis In the OM arm.	Unfunded study—no competing financial interests for the investigators
Perez otero 2022 [16]	3	USA	Median (IRQ) NOM 10.2 (8.5–11.1) surgery 9.7 (7.3–14.4)	39	IV piperacillin/tazobactam for 24–48 h followed by 10 days of oral ciprofloxacin and metronidazole	Surgical appendicectomy	1 month 12 months	Success rate of antibiotics alone, defined as a lack of recurrent appendicitis treated with appendectomy	No funding source for this study
Sajjad 2021 [17]	1	Pakistan	Mean (SD) NOM 9.56 (1.82) Surgery 10.11 (1.83)	180	IV meropenem 10 mg/kg/dose 8 hourly and metronidazole 20 mg/kg/dose 8 hourly for at least 48 h Change to oral ciprofloxacin 10 mg/kg/dose twice daily and metronidazole 20 mg/kg/day two divided doses for another 8 days	Open appendicectomy	6‐month	Failure of nonoperative treatment was defined if Any one of the following is seen: Abscess formation Or complex peri‐appendiceal fluid collection On ultrasonography, the need for surgery (due to worsening of symptoms evaluated by history, physical examination, and repeat USG)	No grant support and financial disclosure

3.1. Study Characteristics

Different approaches to UAA diagnosis were adopted in the included studies. The study by Svensson et al. was the only study to specifiy that UAA diagnosis was made through clinical examination, laboratory tests, and radiological imaging (Ultrasound(US) scan and CT scan were performed when there was diagnostic uncertainty) [18]. Three studies made diagnosis of UAA through clinical examination, while the use of radiological imaging was not clearly stated [19, 20, 21]. Adams et al. reported the number of imaging investigations performed in the Results section [19]. Peretz‐Otero et al. performed US and/or CT to determine perforation or abscess absence and specified the use of abdominal US, CT scan, or both, in the Results section [16]. Hall et al. stated that there was no requirement for diagnostic imaging because this is not routinely used in the clinical diagnosis of UAA in the UK [21]. Sajjad et al. specified that clinical diagnosis of UAA was performed only through the pediatric appendicitis score (PAS) [17, 22].

Variability was also observed in the selection of antibiotic regimens for NOM, including the use of different pharmacological classes. Two studies adopted a scheme based on IV meropenem 10 mg/kg/dose 8 hourly plus metronidazole 20 mg/kg/dose 8 hourly for at least 48 h [17, 18]. Two studies foresee the use of IV piperacillin/tazobactam for the 24–48 h scheme; only Adams et al. specified the dosage (100 mg/kg) [16, 19]. St. Peter et al. reported the adoption of ciprofloxacin and metronidazole, without dosage measures [20]. Hall et al. did not report the administered antibiotics as they were administered by local antimicrobial policy [21].

3.2. Complication‐Free Treatment Success Rate at 30 days

All the included studies reported data on the outcome “complication‐free treatment success rate at 30 days”. A clinically relevant difference emerged in terms of complication‐free treatment success between the two groups (1333 participants; RR 0.90, 95% CI 0.84–0.96; I ² 63%; and moderate certainty of evidence) favoring appendectomy over NOM among the five studies included in the meta‐analysis (Figures 2 and 3). Hall et al. did not report the results in the appendectomy group and was excluded [21].

FIGURE 2.

Complication‐free treatment success rate at 30 days.

FIGURE 3.

Summary of findings’ table.

3.3. Antibiotic Treatment Failure During Hospital Stays and Recurrence of Appendicitis at 6‐Month and 1‐Year Follow‐Up

We collected data on the failure of antibiotics during initial hospital stay, and it is defined as the deterioration of clinical condition or no improvement after 48h in the NOM group, in all the included studies. Data varied from 4% (1/24) reported by Svensson et al. [18] to 30% (8/27) reported by Hall et al. [21].

We extracted data on recurrent appendicitis development in the NOM with antibiotics group up to 6‐month follow‐up from Hall et al. [21] and Sajjad et al. [17]. Hall et al. [21] reported that six (37%) of the 19 participants who initially responded to the antibiotic treatment underwent appendectomy, whereas Sajjad et al. [17] reported 10 cases (11.6%) over 85 participants.

Four included studies reported 1‐year appendicitis recurrence rate that resulted in appendectomy among participants randomized to the NOM group [16, 18, 19, 20]. The percentage of recurrences at 1 year in participants who initially responded to antibiotics varied from 6.25% (1/16) in Peretz‐Otero et al. to 24.1% (22/91) in Adams et al. [16, 19]. Notably, even if Svensson et al. described that 7/22 children who initially responded to antibiotics underwent appendectomy, only one of them had histopathological confirmation of recurrent acute appendicitis; the others underwent surgery due to surgeon or parental discretion without final evidence of acute appendicitis [18].

3.4. Negative Appendectomy Rate

All studies excluding that of Sajjad et al. reported information on NAR, referring to the percentage of appendectomies performed where the appendix was found to be normal upon pathological examination [17].

We included two studies in the analysis investigating NAR (Figures 3, 4). No differences were detected between the two groups (1064 participants; RR 0.61, 95% CI 0.36–1.04; I ² 78%; and very low certainty of evidence) [19, 20].

FIGURE 4.

Negative appendicectomy rate.

Svensson et al. and Peretz‐Otero et al. reported data on NAR only for NOM group participants who were readmitted for appendicitis recurrence (27.3% and 33.3%, respectively) [16, 18]; Hall et al. reported information on NAR only for the surgery group (7.4%) [21].

3.5. Post‐Intervention Complications

All the studies included reported data on post‐intervention complications. For the outcome intraoperative findings of perforated appendicitis, we were able to pool the data for a meta‐analysis.

3.5.1. Intraoperative Findings of Perforated Appendicitis

Three studies were included in the pooled analysis investigating the rate of intraoperative findings of perforated appendicitis [18, 20, 21].

The analysis (Figures 3, 5) revealed no difference between the two groups RR 0.85 (sample size: 950; 95% CI 0.55–1.31; I ² 0%; and moderate certainty of evidence).

FIGURE 5.

Rate of perforated appendicitis.

The study by Adams et al. was not included in the meta‐analysis as the authors did not specify whether the complicated appendicitis reported in 10 of 90 patients (11.11%) in the surgery group was perforated.

3.5.2. Surgical Site Infections

Three studies assessed SSIs occurrence after surgery [19, 20, 21]. St. Peter et al. reported data on SSIs for both patients in the appendectomy and NOM groups readmitted for appendicitis recurrence: 8/394 (2%) and 2/81 (2.5%), respectively, as well as Adams et al.: 3/91 (3.3%) and 2/58 (3.4%), respectively [19, 20]. St. Peters et al. highlighted that 3/8 (37.5%) patients with SSIs developed an abscess, of whom two had perforated appendicitis [20].

Hall et al. referred that 5/27 patients (18.5%) developed SSIs, including fever and abdominal pain (1), inflamed wound site and wound dehiscence (1), wound infection (1), and intra‐abdominal fluid collection/abscess (2). Four patients required additional antibiotic treatment, and one underwent drain insertion and hospitalization. No information was provided for patients in the NOM group that finally underwent appendectomy [21].

3.5.3. Allergic Reaction to Antibiotic Treatment

Allergic reactions to antibiotic treatment were assessed in three of the six included studies. In Adams et al., none of the patients included in both NOM and appendectomy groups underwent complications related to antibiotics [19]. Hall et al. reported two participants in the NOM group with a rush while receiving antibiotics [21], while in St. Peters et al., one participant in the NOM group had an allergic reaction to antibiotics [20].

3.5.4. Other Post‐Intervention Complications

Sajjad et al. did not report information on post‐intervention SSI, but 1/90 (1.11%) patient was readmitted in hospital for adhesive small bowel obstruction [17]. One patient in the surgery group reported suture‐related complications in Hall et al. [21]. Adams et al. reported different post‐intervention complications during the 1‐year follow‐up [19]. Among the participants in the NOM group who had a recurrence and underwent appendectomy after initial response to antibiotics, two had peritonitis, three had abscess, and one had adhesive bowel obstruction. Conversely, only two participants randomized to appendectomy had an abscess. None incurred sepsis [19].

In Peretz‐Otero et al., two patients in the appendectomy group were readmitted to the emergency department, one for constipation and one for pain and wound swelling [16]. St. Peters et al. reported that no patients in both groups had an abscess as a post‐intervention complication [20]. In Svensson et al., during the 1‐year follow‐up, there were no major or minor complications in the two groups.

3.6. Length of Hospital Stays and Returns to Normal Activity

Four of the six included RCTs reported data on LOS as median (IQR) in both treatment groups resulting in a prolonged hospitalization in the NOM with antibiotics group [15, 16, 17, 18, 19, 20]. Svensson et al. [18] and Adams et al. [19] described the LOS in hours. For both the surgery groups, the LOS was 34.5 (16.2–95.0) and 39.32 (25.4–48.78), respectively, while for NOM groups, it was 51.5 (29.9–86.1) and 43.97 (29.00–63.40), respectively. Peretz‐Otero et al. [16] and St. Peters et al. [20] instead reported LOS in days. In the appendicectomy groups, it was 1 (1.0–2.0) and 1 (0.76–1.68) respectively, whereas for NOM groups, it was 2 (1.0–2.0) and 1.25 (0.92–2.09), respectively.

Three studies reported information on the time to return to normal activities [16, 19, 20]. In particular, the median time was 4 (2–5) versus 1 (1–3) for St. Peter et al. [20] and 12 (7.0–14) versus 2 (1.0–10) for Peretz Otero et al. in the comparison between appendectomy and NOM groups, respectively [16] in the comparison between appendectomy and NOM groups, respectively. Adams et al. did not report data per treatment group but for the totality of participants as median days (IQR) 12.00 (7.00–21.00) [19].

3.7. Quality of Life

Peretz‐Otero et al. evaluated the parental Peds QL TM score‐validated scale, highlighting comparable QOL among the two treatment groups, with a median (IQR) of 90.2 (70.1–97.0) in the appendectomy group and 91.3 (82.6–98.9) in the NOM group [16].

3.8. Total Costs

Svensson et al. reported the median cost (range) expressed in SEK—official Sweden currency—of initial inpatient stay for appendicectomy (45,805 SEK, 33,042–94,638) and for NOM with antibiotics (30,732 SEK, 18,980–63,863). The median total cost was 45,805 SEK (33,042–97,638) for appendectomy and 34,587 SEK (19,120–146,55) for NOM [18].

3.9. Risk of Bias

We assessed RoB for complication free‐treatment success at 30 days and NAR.

For the primary outcome, four of the six RCTs were judged at low risk of bias [18, 19, 20, 21], one at high risk [16], and one with some concern [17] according to the ROB2 tool [13].

All the included studies generated random sequence adequately, except for Peretz‐Otero et al., [16] which did not report information on allocation concealment, resulting in some concern of the randomization process.

Considering the type of treatment assigned to the appendectomy group, none of the included studies attempted at blinding participants and surgeons delivering the intervention. Otherwise, Hall et al. specified that the study statistician performing the data analysis was blinded to treatment allocation using coded data [21]. All but two [16, 17] of the RCTs included were considered at low risk of bias of deviation of intended intervention as according to the protocol, patients in the NOM group in which antibiotic treatment failed, underwent appendectomy. In Peretz‐Otero et al., conversely, some participants were moved from the NOM group to the appendectomy group after randomization before the treatment begun [16]. In Sajjad et al., no clear information emerged about possible deviation from the intended intervention [17].

St. Peters et al. was judged at low risk of bias, while Adams et al. was judged at some concern due to the bias during selection of the reported results for the secondary outcome NAR, as this outcome was not scheduled to be assessed [19, 20].

4. Discussion

This review includes six RCTs aimed at comparing NOM with antibiotics versus appendectomy in children and adolescents.

Our findings suggest a statistically significant benefit for appendectomy over NOM in terms of a complication‐free treatment success rate at 30 days. However, the relative effect estimate (RR 0.90) indicates that NOM with antibiotics may still achieve favorable outcomes in a substantial proportion of patients. No significant differences were observed between groups in terms of NAR and intraoperative findings of perforated appendicitis.

4.1. Strength of the Review

A major strength of this review is the inclusion of only RCTs with pediatric‐specific populations, thereby addressing a key limitation of prior reviews that combined data from mixed‐age or mixed‐study designs.

Another strength is the GRADE approach for the certainty of evidence assessment [14]. The GRADE evaluation on the primary outcome was rated moderate due to the high risk of bias of one of the included studies for this outcome. The GRADE evaluation on the secondary outcome of NAR was rated low due to risk of bias, imprecision, and inconsistency, whereas the secondary outcome of rate of perforated appendicitis was rated high despite downgraded for imprecision. However, given the small magnitude of the effect in terms of these three outcomes, it is likely that additional studies could change the direction of the pooled estimates.

4.2. Limitation of the Included Studies

NOM failures could be attributed mainly to a wrong UAA diagnosis. Indeed, one of the most relevant limitations observed in our study concerns the diagnostic approaches' heterogeneity in identifying uncomplicated cases across the included RCTs. Although all trials claimed to use clinical diagnostic criteria, only one RCT explicitly detailed the diagnostic process incorporating clinical evaluation and laboratory testing [18], while the three newer studies relied heavily, or exclusively, on clinical judgment, without clear documentation of whether imaging was employed systematically or only as an adjunct [19, 20, 21]. This variability compromises internal and external validity, especially given that clinical presentation of acute appendicitis can be ambiguous in younger patients, leading to over‐ and/or under‐diagnosis [23].

Another issue to consider is the standardization and safety of the diagnostic process. Data from the 2025 World Society of Emergency Surgery (WSES) guidelines suggest that US should represent the first‐line imaging test for children and adolescents with suspected acute appendicitis, reserving low‐radiation‐dose CT for cases in which US is nondiagnostic or clinical suspicion remains high [24].

4.2.1. The Main Limitation of NOM With Antibiotics is the Possibility of Appendicitis Recurrence, as it Occurred

From the included studies, it was reported that appendicitis recurrence could affect up to 37% of patients treated with NOM within 12 months after antibiotic treatment (range 4–37), and several patients underwent interval elective appendicectomy. However, many recurrences resulted in a negative appendectomy, raising concerns about the appropriateness of surgery. This is further complicated by the lack of standardized criteria to define recurrence, as patients with nonspecific abdominal pain are often categorized as having recurrent appendicitis and may undergo surgery without definitive diagnostic confirmation.

Another knowledge gap concerns the outcomes' underreporting of patients who initially received NOM but ultimately required appendectomy due to recurrence or treatment failure. While recurrence rate and NAR for these patients were variably reported, there was little to no information on the nature and frequency of complications encountered after appendectomy for failed NOM. This is a missed opportunity, as these distinct subgroups may carry different risk profiles. Future RCTs should systematically report postoperative complication rates in both immediate and late appendectomy groups. Such data are essential to inform shared decision‐making discussions, especially in scenarios where parents or clinicians may opt for initial NOM with antibiotics under the assumption that surgical risks remain equivalent regardless of timing.

Another shortcoming is the near‐complete absence of patient‐reported outcome measures and caregiver‐related social or economic impacts. Only one study included a validated QoL assessment, and even that was limited to a single follow‐up timepoint [16]. No study reported data on emotional well‐being, or functional recovery from the perspective of a child, despite increasing emphasis on incorporating such measures into pediatric surgical research. In economic terms, parental absenteeism also translates into real costs for families and society.

NOM with antibiotics appears associated with faster recovery and return to daily activities. However, the need for outpatient follow‐up and the theoretical risk of recurrence might influence parental time away from work. In the absence of data, this remains only speculative. Within this context, shared decision‐making is a fundamental aspect of pediatric care, particularly when multiple management strategies offer comparable safety and efficacy, as for UAA. Engaging parents in a transparent discussion about the potential benefits and risks of NOM (recurrence, possibility of delayed surgery, postoperative complications, and impact on daily activities) can help align treatment decisions with family values and preferences. Most parents value the opportunity to participate in treatment decisions and may choose antibiotics when provided with evidence‐based information and clinical support [25]. Furthermore, pediatric patients and caregivers can process information in the acute care setting and effectively participate in an informed shared decision‐making process around the need for surgery [26, 27]. In 2020, a group of researchers conducted a survey focused on parental preferences in the treatment of UAA comparing surgery to NOM and their views on research participation showing that when given the choice, most prefer NOM. An even larger proportion finds randomization a barrier to participation in research. These findings have implications on protocols study design in general and specifically in UAA trials in children [28].

4.3. Limitation of the Review

We searched the bibliographic databases from inception to March 2025, but we did not update the whole search again at the end of the review process. We did not contact the authors of the trials we included to seek for additional information.

The included RCTs showed substantial heterogeneity in outcome definitions. Although we attempted to extract data from outcomes that differed in definition and follow‐up duration, these outcomes were intended to measure similar end points.

Differences in diagnostic pathways across studies may introduce misclassification bias, particularly in distinguishing complicated versus uncomplicated appendicitis and may contribute to the observed heterogeneity across studies (I ² ranging from 63% to 78%).

The apparent differences between our findings and those of previous systematic reviews can be largely explained by methodological heterogeneity, particularly in the definition of primary outcomes. Our study used a composite, patient‐centered end point (complication‐free treatment success at 30 days), which is more stringent than the outcomes adopted in most prior meta‐analyses. Among the studies cited, only the meta‐analysis by Brucchi et al. [9] adopted a similar definition and reported a direction of effect favoring surgery, although not statistically significant (likely due to the limited number of included RCTs and the absence of more recent large trials). Other meta‐analyses used fundamentally different primary outcomes, such as treatment failure or success at 1 year, which are more heavily influenced by recurrence (an outcome that cannot occur in one of the two treatment branches) [29, 30]. Furthermore, our analysis incorporates the most recent high‐quality RCTs, including large multicenter studies, which were not consistently included in earlier syntheses. This likely contributes to the observed statistical difference at 30 days.

In light of these considerations, antibiotic therapy could represent a feasible treatment option for UAA in children and adolescents, although, in terms of complication‐free treatment success, appendectomy remains the gold standard.

Future RCTs should focus on standardizing antibiotic regimens, evaluating the definitive role of antibiotics in the treatment of UAA, optimizing the diagnostic process, and incorporating longer term and patient‐centered outcomes, including HRQoL and return to baseline functioning.

Studies conducted in low‐ and middle‐income countries are needed to assess generalizability and equity of access to both surgery and medicines.

Author Contributions

Paulina Salminen: supervision, validation. Eleonora Allocati: methodology, data curation, conceptualization, writing – review and editing, writing – original draft, formal analysis. Chiara Gerardi: writing – review and editing, writing – original draft, conceptualization, methodology, supervision, data curation. Marco Ceresoli: validation, writing – review and editing, supervision. Benedetta Starinieri: formal analysis, methodology, writing – original draft, data curation. Mauro Podda: conceptualization, writing – review and editing, supervision, validation, methodology.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

Salminen P reported receiving financial support from the Sigrid Jusélius Foundation; receiving research grants from the Academy of Finland and the European Research Council; being a lecturer for Johnson & Johnson, BD, Novo Nordisk; being a member of the data safety monitoring board of the Best RCT (Sweden) and the Magnet Study. All other authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The main information are reported in the manuscript.