Introduction

A perianal abscess (PA) is a common infectious disease in infants that develops from infection in abnormally deep crypts of Morgagni1, with a strong male predominance2. The incidence was highest in the first 3 months of age, decreased gradually until 1 year of age3. Presently, the optimal management of PA in infant has not yet been established. The most common treatment methods include incision and drainage (ID)4,5,6,7,8,9,10,11,12,13,14,15, incision and drainage with primary fistulotomy (IDF)16,17,18,19,20,21,22,23,24,25,26,27, and conservative management (CM)2,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43. Among them, ID is currently the preferred treatment choice. IDF is also a prevalent treatment modality, exhibiting higher cure rates. Since 1998, PA in infants has been recognized as a self-limiting disease30, and CM has been proven to be effective. Due to a lack of large sample size and long-term follow-up data, there is currently no universally recognized optimal treatment method. Further research is needed to address these issues.

Our previous study compared the outcomes of ID versus IDF for PA in neonates. We found both to be safe and effective treatment modalities, with IDF demonstrating superior efficacy44. In our latest study, we found that CM resulted in a long-term cure rate of 90.8%3. In the last four years, our understanding of PA in infants has continuously deepened. What is the optimal management?

To the best of our knowledge, no study has directly compared the outcomes of ID, IDF, and CM for PA in infants. This study aimed to compare the efficacy and safety of ID, IDF, and CM for infants with PA.

Materials and methods

This study is a retrospective cohort study, with the study population consisting of infants with PA who were treated at our center from September 2014 to December 2020 (aged less than 1 year). Before 2023, the treatment for PA in infants was mainly surgical management at our hospital. Surgeons in different departments have different discretion for specific surgical procedure. Infants with PA can be treated either in the department of pediatric surgery (ID) or anorectal surgery (IDF). The decision of whether an infant visited the department of pediatric surgery or the colorectal surgery is determined by the parents’ awareness and preferences, rather than the severity of the infant’s condition. Infants who underwent IDF in the department of anorectal surgery were included in the IDF group, while those who underwent ID in the department of pediatric surgery were included in the ID group. Infants whose parents refused surgery and opted for conservative treatment were included in the CM group. The grouping in this cohort was not artificially intervened; rather, it was naturally formed. Patients in the ID and IDF groups received intravenous antibiotics routinely (Cefathiamidine in the ID group and Ceftazidime in the IDF group) for 3–5 days.

It’s important to note that our previous study3 shares the same patient population as the current study, with the sample of patients undergoing CM being identical. The difference lies in the criteria for treatment failure; in the previous study, treatment failure included new-onset abscesses, whereas in the current study, new-onset cases are discussed separately and not included in the statistics for treatment failure.

This study has received approval from the Affiliated Hospital of Jining Medical University Institutional Review Board. All methods were performed in accordance with the relevant guidelines and regulations. Informed consent was obtained from all participants’ legal guardians.

Incision and drainage

Under local anesthesia, a small incision was carefully created at the apex of the abscess. Subsequently, the internal pockets within the cavity were meticulously disrupted through gentle probing using a hemostatic forcep.

Incision and drainage with primary fistulotomy

Immediately following ID, a fistulotomy was routinely performed with the patient positioned in the left lateral decubitus position under conscious sedation and local anesthesia. The critical step was identifying the internal opening. A fine probe was carefully advanced from the abscess cavity to the affected anal crypt, after which the fistula tract was unroofed and opened using diathermy. If the internal opening could not be probed, the corresponding internal opening, located at the center of the abscess, was identified and laid open.

Conservative management

Before 2023, surgical intervention was the standard approach for managing PA in infants at our center. The patients were treated conservatively because their parents refused surgery. Conservative treatment consisted of sitz baths with traditional Chinese medicine, hygiene, erythromycin ointment, and mupirocin ointment, et al. Antibiotics (one course) were not routinely used. To fully observe the effects of CM, patients who underwent surgical management within 2 months of CM were excluded.

Diagnostic criteria

The diagnosis was made using the International Classification of Diseases (ICD) criteria. Diagnosis of PA and anal fistula is based on medical history and physical examination. Key points for diagnosing PA include redness, swelling, and pain around the anus, with palpable firm masses in the perianal area.

Definitions

Treatment Failure: This encompasses the anal fistula formation and recurrence.

Fistula Formation: The formation of a fistula opening in the perianal skin with a tract between the anal canal and rectum.

Recurrence: Refers to the reappearance of PA in the original location after the abscess was cured.

Cure: There is neither recurrence nor fistula formation.

New-onset abscess: Formation of a PA in a location outside the primary lesion site.

Inclusion criteria

The study population consists of infants with PA (aged less than 1 year) who were first diagnosed and treated at our outpatient department from August 2014 to December 2020.

Exclusion criteria

Infants with PA who received treatment at another hospital before visiting our outpatient department, infants treated conservatively for less than 2 months, infants with disease duration exceeding three months at the time of consultation resulting in the formation of an anal fistula, as well as patients lost to follow-up or with incomplete medical records, were excluded from the study. Infants with inflammatory bowel disease and systemic diseases were excluded from our study.

Follow-Up

Short-term follow-up was performed weekly using outpatient evaluations in the first three weeks, and monthly thereafter. Long-term follow-up data were predominantly acquired through telephone interviews or WeChat (a Chinese multi-purpose messaging, social media app developed by Tencent. It not only allows users to send text messages, pictures, and videos, but also to make voice and video calls).

Data collection

The collected data encompassed baseline characteristics, duration of follow-up, fistula formation, recurrence of abscesses, incidence of new-onset abscesses, cure time, and fecal incontinence. The abscess size was recorded as the longest abscess diameter, excluding the depth.

Statistical analysis

Quantitative variables were presented as medians accompanied by interquartile ranges (IQR). Categorical variables were expressed as frequency (percentages). As the quantitative variables did not meet the assumption of normal distribution, Kruskal–Wallis H-test or the Mann–Whitney U-test (length of stay, and inpatient costs) were utilized. Chi-square and Fisher’s exact tests were used for contingency table analysis. Bonferroni correction was applied to post hoc pairwise comparisons. A p-value of less than 0.05 was considered to indicate statistical significance. All statistical analyses were conducted using SPSS software, version 23.0.

Results

In total, 772 infants with PA (< 1 year) were identified from 2014 to 2020. 597 patients were included in the statistical analysis according to the inclusion and exclusion criteria. A flow chart showing patient screening is shown in Fig. 1.

Fig. 1
Fig. 1
Full size image

The flowchart of patient inclusion. PA perianal abscess.

Table 1 summarizes the comparison of baseline characteristics including gender, age at presentation, number of abscesses, abscess size, location of abscessed, length of stay, inpatient costs, and follow-up among the three groups of patients. The median follow-up period was 5.8 years (range 3–9.4 years). There were no statistically significant differences among the three groups in terms of follow-up time. Except for abscess size, length of stay, and inpatient costs, the baseline characteristics of the three groups were similar, with no statistically significant differences observed (p > 0.05).

Table 1 Baseline characteristics of infants with PA. PA perianal abscess, ID incision and drainage, IDF incision and drainage with primary fistulotomy, CM conservative management, IQR interquartile range, RMB renminbi, the currency of China. *Fisher’s exact probability method; **Kruskal–Wallis H-test.
Full size table

Outcomes of the three groups are shown in Table 2. The cure rate in the IDF group was 97.8% (276/282), significantly higher than that in the ID group (82.1%, 133/162; p < 0.001) and the CM group (80.4%, 123/153; p < 0.001). Moreover, there was no significant statistical difference in cure rates between the ID group and the CM group. Pairwise comparisons with Bonferroni correction are shown in Table 3.

Table 2 Outcomes of the three groups. ID incision and drainage, IDF incision and drainage with primary fistulotomy, CM conservative management. *Fisher’s exact probability method.
Full size table
Table 3 Post hoc pairwise comparisons with Bonferroni correction for outcomes of the three groups. ID incision and drainage, IDF incision and drainage with primary fistulotomy, CM conservative management. A cure rate; b rate of fistula formation; c rate of abscess recurrence; d rate of new-onset abscess. *Fisher’s exact probability method.
Full size table

In the ID group, of the 29 treatment failures, 22 were cured by surgery and 7 by CM alone. Among the 6 treatment failures in the IDF group, 5 were cured with conservative treatment and 1 underwent surgery. In the CM group, out of 30 treatment failures, 10 underwent surgery and 20 continued with CM; 17 were cured and 3 remained uncured.

During the treatment process, new abscess formation occurred in all three groups of patients. Statistical analysis showed a significant difference in the rate of new-onset abscesses across the three groups. However, post hoc pairwise comparisons with Bonferroni correction did not reveal any significant differences between any two groups (see Table 3). The age range of patients with newly developed abscesses is from 15 to 281 days. The time from the initial onset of symptoms to the development of the new abscess ranges from 3 to 8 years, with the majority occurring within 2 months. Among the 11 new cases in the ID group, 8 cases underwent surgery, and 3 cases opted for conservative treatment. In the IDF group, out of the 6 cases of new abscess formation, 5 cases opted for conservative treatment, and 1 case underwent surgery. In the CM group, out of the 10 new cases, 4 cases underwent surgery, and 6 cases opted for conservative treatment. All the mentioned patients achieved cure.

No severe complications were found in the ID or CM groups. The IDF group had 3 cases of sepsis, including 1 with a concurrent central nervous system infection. The preoperative white blood cell counts for the three patients were 15.1, 21.3, and 35.6 × 109/L, respectively. No fecal incontinence was observed in any of the patients.

Discussion

Our study investigated the efficacy of different treatment strategies for pediatric PA in infants under 1 year of age. We observed a total of 772 infants with PA in the outpatient setting over a period of 6 years, ultimately enrolling 597 patients in our study after exclusions and losses to follow-up. We achieved a high follow-up rate of 87.3%, with a median follow-up period of 5.8 years.

In our study, all three treatment methods were safe and effective. However, the efficacy of conservative treatment exceeded our expectations, with a cure rate of 80.4% (123/153) 2 months after initial diagnosis and a cure rate of 91.5% (140/153) at the long-term follow-up. This suggests that PA in infants may be a self-limiting condition, which is consistent with previous studies30,36,41. A study indicates that the cure rates for abscesses that naturally rupture and drain are the same as those for surgically drained abscesses32. For patients whose abscesses had already ruptured at the time of consultation, further ID did not provide additional benefits, making CM a better choice. If there is no significant progression during the treatment process, extending the duration of CM can improve the cure rate.

In the ID and IDF groups, a total of 59 patients experienced surgical treatment failure, of which 29 (50%) were cured with CM. This demonstrates that CM can be used not only for first-time PA, but also for patients who have failed surgical treatment, potentially reducing the trauma caused by secondary surgeries and anesthesia.

The cure rate in the IDF group was 97.8% (276/282), significantly higher than that in the ID group and the CM group. We believe that excising the fistula, incising the internal opening, or removing the corresponding anal crypt ensures more adequate drainage and mitigates factors that contribute to recurrence and the formation of fistulas, thereby enhancing the cure rate. This is fully consistent with previous literature, which showed no recurrences19,21 or a recurrence rate of less than 15%17,23 in patients who underwent fistulotomy at the time of initial drainage. Based on these findings, we recommended that ID and simultaneous laying open of any coexisting fistulous tract should be performed for PA in neonates44.

It is worth noting that this was our perspective in 2022 (advocating IDF), which is inappropriate. Patients who underwent IDF require anesthesia, are subjected to surgical trauma, and face a potential risk of postoperative complications. Therefore, in spite of a significant cure rate, IDF has ceased to be the mainstream treatment approach at our hospital in 2023.

After 2022, through a retrospective study3, we discovered that infant PA may be a self-limiting condition. CM is effective and safe for PA in infants, achieving a cure rate of over 80% while incurring minimal treatment costs (without severe adverse effects). Since the majority of patients achieved cure with CM, avoiding the risks of anesthesia and surgical trauma, we now recommend CM as the first-line approach for infant PA. Starting in 2023, we initiated a prospective clinical study (ChiCTR2300077406) to evaluate the efficacy and safety of CM for PA in infants. To date, we have enrolled 30 patients, and the preliminary results of our study provide initial evidence supporting the effectiveness of CM.

In the IDF group, three severe complications occurred, including two cases of sepsis and one case of sepsis with a concurrent central nervous system infection. All three patients had multiple abscesses larger than 2 cm and suffered from severe diarrhea and watery stools. In fact, the sepsis did not necessarily result from surgery. We hypothesize that sepsis may be associated with leukocytosis, extensive infection, transient intraoperative entry of purulent fluid into the bloodstream, and postoperative absorption of watery stool through the open wound. Fortunately, after intensive treatment, all three patients recovered well.

In the IDF group, 6 patients (2.2%, 6/282) failed surgery; 5 were cured with conservative treatment and 1 underwent surgery. CM was the main treatment for those who failed IDF. This is mainly because secondary surgeries can potentially lead to medical disputes. In most cases, unless the abscess is too large, the symptoms of anal fistula recur frequently, or the patient’s parents strongly insist on surgery, we prefer to adopt a conservative approach. In fact, out study showed that CM is effective in patients who experience failure following IDF surgery.

ID remains the most common treatment approach for pediatric abscesses, although the recurrence rate and the likelihood of fistula formation vary according to the literature. In our study, the cure rate in the ID group was 82.0% (133/162), consistent with our previous data44 on neonatal PA. Given the similar effectiveness of ID and CM, we prefer CM for initial abscess incidents in infants. Since 2023, we have prioritized conservative treatment as the primary approach (not only due to parental refusal for surgery), moving away from the previously preferred IDF method. Only a minority of patients are suitable for surgical resection.

Determining the appropriate timing for surgical intervention remains a challenge with no universally accepted standard. Based on our clinical experience, we suggest that surgery should be considered under the following circumstances: The patient exhibits abnormal irritability or inconsolable crying; Signs of sepsis are present; The patient has a fever exceeding 38.5 °C; The abscess cavity measures larger than 3 cm in diameter; There are more than three abscesses; The patient does not achieve complete cure despite undergoing several months of CM; The parents of the patient strongly request surgical intervention.

The decision to proceed with surgery should also take into account the preferences of the patient’s parents, as this is a shared decision-making process. Effective communication is essential, especially since CM is not guaranteed to be fully effective. Providing comprehensive information can help mitigate potential conflicts between healthcare providers and patients’ families.

An interesting and unique finding in our study is the asymmetrical occurrence of PA among the 597 pediatric patients; right-sided abscesses (at the 9 o’clock position) occurred at a higher rate of 53.9% (322/597) compared to the left side (at the 3 o’clock position) at 36.3% (217/597) (P < 0.05), with bilateral occurrences at 7.2% (43/597) and other locations at 2.5% (15/597). This is the first time such a phenomenon has been observed and has not been previously reported. We speculate that this may be related to anatomical differences in the anal glands, but the reason for this finding is not yet understood and requires further study.

Our data show that among the 48 patients who developed fistulas, the rate of new abscess formation was 18.7% (9/48), which is higher than the overall new abscess rate of 4.5% (27/597, p < 0.001). It is also significantly higher than the new abscess rate of 3.3% (18/549, p < 0.001) in patients who did not develop fistulas. The reasons for this correlation are not yet understood and may be related to the longer disease course, repeated purulent discharges irritating the anal skin, or other unclear factors. Therefore, in patients with fistulas who continued with CM, attention should be paid to new abscess formation.

In the CM group, out of the 10 new cases, 6 cases opted for conservative treatment. Some patients developed new abscesses during the course of CM, yet their parents continued to reject surgical intervention. Through follow-up, we found that CM was also effective for managing newly developed abscesses. Among the three groups of patients, a total of 27 cases of new abscess formation occurred. Of these, 17 patients chose surgery and 10 opted for CM as their parents refused surgery. Ultimately, all 27 patients with new abscesses were cured. This may be because that PA in infants may be a self-limiting condition, suggesting that CM is effective, with surgery being best avoided whenever possible.

Fecal incontinence can be a potential risk associated with fistulotomy due to the damage of the sphincter. In fact, fistulotomy is not associated with the risk of fecal incontinence, as the fistulae in infants are almost always sphincteric17,19,23,26,27,45. Only two patients have been reported to have developed fecal incontinence following surgery21. Thus, the risk of fecal incontinence following the IDF procedure is extremely low, making it a negligible concern in clinical practice.

Our study’s large sample size and long-term follow-up add reliability, but it is important to note that our study is a single-center retrospective study, and its conclusions are somewhat limited. What’s more, the abscess size was different between the three groups. It should be noted that the 153 patients in the CM group were treated conservatively because their parents refused surgical intervention, rather than due to the small size of the abscesses. In fact, 67 patients (43.8%) had abscesses larger than or equal to 1 cm. We must acknowledge that the selection of patients for CM may be subject to bias; however, the impact of this bias is likely minimal. Furthermore, we are currently conducting a prospective cohort study on CM for infants with PA, which will help to significantly reduce this bias.

Conclusions

Our data suggest that PA in infancy is likely a self-limiting disease, with most patients responding well to extended CM. For patients who develop fistulas following failed surgical treatment, CM is also recommended. Early surgical management should be avoided as the first choice for PA in infants.