What are the benefits and risks of different treatments for bacterial folliculitis and boils (inflammation of the skin around hairs)?

Why is this question important?

Bacterial folliculitis is an inflammation of the tiny pockets in our skin from which hairs grow (hair follicles). It occurs when bacteria (tiny organisms not visible with the naked eye) infect hair follicles. Bacterial folliculitis typically causes red swelling, with or without a small blister that contains pus.

Without treatment, bacterial folliculitis may progress to hard and painful lumps filled with pus, known as boils. These cover several hair follicles, and affect the skin around them.

Bacterial folliculitis and boils affect people worldwide, and have an important negative impact on quality of life. Infections typically:

- cause unsightly infections on parts of the body visible to others (such as the face and neck); or

- develop where skin rubs, causing discomfort and pain (such as armpits and buttocks).

A range of treatment options for bacterial folliculitis and boils is available. These include:

- antibiotics (medicines that fight bacterial infections). These can be applied to part of the body (locally) in the form of creams (topical antibiotics); or they can be taken by mouth (orally) or given as injections, to treat the whole body (systemic antibiotics);

- antiseptics (chemicals applied to the skin to fight infections caused by micro-organisms, such as bacteria);

- light therapy; and

- surgery, for example, doctors may make a small cut (incision) in the skin to allow pus to drain out.

To find out which treatments work best for bacterial folliculitis and boils, we reviewed the evidence from research studies.

How did we identify and evaluate the evidence?

First, we searched for randomised controlled studies, in which people were randomly put into one of two or more treatment groups. This makes it less likely that any differences between treatments were actually due to differences in the people who received them (rather than the treatments themselves, which is what we wanted to find out).

We then compared the results, and summarised the evidence from all the studies. Finally, we rated our confidence in the evidence, based on factors such as study methods and sizes, and the consistency of findings across studies.

What did we find?

We found 18 studies that involved a total of 1300 people. People were followed-up for between one week and three months. Studies were set in Asia, Europe and America. Only three studies reported information about funding: non-profit organisations funded one study, and pharmaceutical companies funded two studies.

The studies compared:

- different oral antibiotics (11 studies);

- different topical antibiotics (2 studies);

- different treatments for wound care after boil incision (2 studies);

- different traditional Chinese medicines (1 study);

- co-trimoxazole (antibiotics) with, and without, 8-methoxypsoralen (a light-sensitising treatment) followed by exposure to sunlight (1 study); and

- penicillin (an antibiotic) with, and without, fire cupping (a form of traditional Chinese medicine) after surgery (1 study).

We found no studies that evaluated antiseptics or investigated quality of life or recurrence of bacterial folliculitis or boils.

Here we report the findings from four comparisons of different oral antibiotics.


The evidence from studies that investigated how successfully different oral antibiotics cured bacterial folliculitis and boils suggests that:

- there is probably little to no difference between cefditoren pivoxil and cefaclor (1 study, 93 people);

- there may be little to no difference between cefdinir and cephalexin (1 study, 74 people).

The few studies available did not provide sufficiently robust information to determine if:

- cefadroxil is better or worse than flucloxacillin (1 study, 41 people); or

- azithromycin is better or worse than cefaclor (2 studies, 31 people).

Severe adverse events (such as fever or vomiting)

The evidence from studies that compared frequencies of severe adverse events suggests there may be little to no difference between:

- cefadroxil and flucloxacillin (1 study, 651 people);

- cefdinir and cephalexin (1 study, 391 people); and

- cefditoren pivoxil and cefaclor (1 study, 150 people).

We do not know if azithromycin is associated with more, or fewer, severe adverse events than cefaclor. This is because studies provided insufficiently robust information (2 studies, 274 people).

Minor adverse events (such as feeling thirsty or dizzy)

The evidence from studies that compared frequencies of minor adverse events suggests there is probably little to no difference between:

- cefadroxil and flucloxacillin (1 study, 651 people); and

- cefditoren pivoxil and cefaclor (1 study, 150 people).

We do not know whether there are more, or fewer, minor adverse events associated with:

- cefdinir or cephalexin (1 study, 391 people); or

- azithromycin or cefaclor (2 studies, 274 people).

This is because studies reported insufficiently robust information.

What does this mean?

The limited evidence available does not suggest that any one oral antibiotic is better than another for treating bacterial folliculitis and boils.

The comparative benefits and risks of other treatments such as antiseptics or light therapy are unclear, because too few studies have investigated this.

How up-to-date is this review?

The evidence in this Cochrane Review is current to June 2020.

Authors' conclusions: 

We found no RCTs regarding the efficacy and safety of topical antibiotics versus antiseptics, topical versus systemic antibiotics, or phototherapy versus sham light for treating bacterial folliculitis or boils. Comparative trials have not identified important differences in efficacy or safety outcomes between different oral antibiotics for treating bacterial folliculitis or boils.

Most of the included studies assessed participants with skin and soft tissue infection which included many disease types, whilst others focused specifically on folliculitis or boils. Antibiotic sensitivity data for causative organisms were often not reported. Future trials should incorporate culture and sensitivity information and consider comparing topical antibiotic with antiseptic, and topical versus systemic antibiotics or phototherapy.

Read the full abstract...

Bacterial folliculitis and boils are globally prevalent bacterial infections involving inflammation of the hair follicle and the perifollicular tissue. Some folliculitis may resolve spontaneously, but others may progress to boils without treatment. Boils, also known as furuncles, involve adjacent tissue and may progress to cellulitis or lymphadenitis. A systematic review of the best evidence on the available treatments was needed.


To assess the effects of interventions (such as topical antibiotics, topical antiseptic agents, systemic antibiotics, phototherapy, and incision and drainage) for people with bacterial folliculitis and boils.

Search strategy: 

We searched the following databases up to June 2020: the Cochrane Skin Specialised Register, CENTRAL, MEDLINE, and Embase. We also searched five trials registers up to June 2020. We checked the reference lists of included studies and relevant reviews for further relevant trials. 

Selection criteria: 

We included randomised controlled trials (RCTs) that assessed systemic antibiotics; topical antibiotics; topical antiseptics, such as topical benzoyl peroxide; phototherapy; and surgical interventions in participants with bacterial folliculitis or boils. Eligible comparators were active intervention, placebo, or no treatment.

Data collection and analysis: 

We used standard methodological procedures expected by Cochrane. Our primary outcomes were 'clinical cure' and 'severe adverse events leading to withdrawal of treatment'; secondary outcomes were 'quality of life', 'recurrence of folliculitis or boil following completion of treatment', and 'minor adverse events not leading to withdrawal of treatment'. We used GRADE to assess the certainty of the evidence.

Main results: 

We included 18 RCTs (1300 participants). The studies included more males (332) than females (221), although not all studies reported these data. Seventeen trials were conducted in hospitals, and one was conducted in clinics. The participants included both children and adults (0 to 99 years). The studies did not describe severity in detail; of the 232 participants with folliculitis, 36% were chronic. At least 61% of participants had furuncles or boils, of which at least 47% were incised. Duration of oral and topical treatments ranged from 3 days to 6 weeks, with duration of follow-up ranging from 3 days to 6 months. The study sites included Asia, Europe, and America. Only three trials reported funding, with two funded by industry.

Ten studies were at high risk of 'performance bias', five at high risk of 'reporting bias', and three at high risk of 'detection bias'.

We did not identify any RCTs comparing topical antibiotics against topical antiseptics, topical antibiotics against systemic antibiotics, or phototherapy against sham light. Eleven trials compared different oral antibiotics.

We are uncertain as to whether cefadroxil compared to flucloxacillin (17/21 versus 18/20, risk ratio (RR) 0.90, 95% confidence interval (CI) 0.70 to 1.16; 41 participants; 1 study; 10 days of treatment) or azithromycin compared to cefaclor (8/15 versus 10/16, RR 1.01, 95% CI 0.72 to 1.40; 31 participants; 2 studies; 7 days of treatment) differed in clinical cure (both very low-certainty evidence). There may be little to no difference in clinical cure rate between cefdinir and cefalexin after 17 to 24 days (25/32 versus 32/42, RR 1.00, 95% CI 0.73 to 1.38; 74 participants; 1 study; low-certainty evidence), and there probably is little to no difference in clinical cure rate between cefditoren pivoxil and cefaclor after 7 days (24/46 versus 21/47, RR 1.17, 95% CI 0.77 to 1.78; 93 participants; 1 study; moderate-certainty evidence).

For risk of severe adverse events leading to treatment withdrawal, there may be little to no difference between cefdinir versus cefalexin after 17 to 24 days (1/191 versus 1/200, RR 1.05, 95% CI 0.07 to 16.62; 391 participants; 1 study; low-certainty evidence). There may be an increased risk with cefadroxil compared with flucloxacillin after 10 days (6/327 versus 2/324, RR 2.97, 95% CI 0.60 to 14.62; 651 participants; 1 study; low-certainty evidence) and cefditoren pivoxil compared with cefaclor after 7 days (2/77 versus 0/73, RR 4.74, 95% CI 0.23 to 97.17; 150 participants; 1 study; low-certainty evidence). However, for these three comparisons the 95% CI is very wide and includes the possibility of both increased and reduced risk of events. We are uncertain whether azithromycin affects the risk of severe adverse events leading to withdrawal of treatment compared to cefaclor (274 participants; 2 studies; very low-certainty evidence) as no events occurred in either group after seven days.

For risk of minor adverse events, there is probably little to no difference between the following comparisons: cefadroxil versus flucloxacillin after 10 days (91/327 versus 116/324, RR 0.78, 95% CI 0.62 to 0.98; 651 participants; 1 study; moderate-certainty evidence) or cefditoren pivoxil versus cefaclor after 7 days (8/77 versus 5/73, RR 1.52, 95% CI 0.52 to 4.42; 150 participants; 1 study; moderate-certainty evidence). We are uncertain of the effect of azithromycin versus cefaclor after seven days due to very low-certainty evidence (7/148 versus 4/126, RR 1.26, 95% CI 0.38 to 4.17; 274 participants; 2 studies). The study comparing cefdinir versus cefalexin did not report data for total minor adverse events, but both groups experienced diarrhoea, nausea, and vaginal mycosis during 17 to 24 days of treatment. Additional adverse events reported in the other included studies were vomiting, rashes, and gastrointestinal symptoms such as stomach ache, with some events leading to study withdrawal.

Three included studies assessed recurrence following completion of treatment, none of which evaluated our key comparisons, and no studies assessed quality of life.