Intervention Review

You have free access to this content

Pin site care for preventing infections associated with external bone fixators and pins

  1. Anne Lethaby1,*,
  2. Jenny Temple2,
  3. Julie Santy-Tomlinson3

Editorial Group: Cochrane Wounds Group

Published Online: 3 DEC 2013

Assessed as up-to-date: 5 SEP 2013

DOI: 10.1002/14651858.CD004551.pub3


How to Cite

Lethaby A, Temple J, Santy-Tomlinson J. Pin site care for preventing infections associated with external bone fixators and pins. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No.: CD004551. DOI: 10.1002/14651858.CD004551.pub3.

Author Information

  1. 1

    University of Auckland, Obstetrics and Gynaecology, Auckland, New Zealand

  2. 2

    University of Plymouth, Faculty of Health & Social Work, Drake Circus, Plymouth, UK

  3. 3

    University of Hull, Faculty of Health & Social Care, Hull, East Yorkshire, UK

*Anne Lethaby, Obstetrics and Gynaecology, University of Auckland, Private Bag 92019, Auckland, 1142, New Zealand. a.lethaby@auckland.ac.nz.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 3 DEC 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Camathias 2012

MethodsSingle centre parallel group study.

Power calculation for sample size, based on results from Davies 2005 study - proportions of infection in 2 groups with different pin site care (80% power and alpha value of 0.05)

Patients were not randomised; pins were randomised.

No of pins randomised = 204

No of pins analysed = 204

No of patients = 56

Source of funding: nil


ParticipantsInclusion criteria: patients treated with a standard external fixator (ExFix)

Exclusion criteria: application of ExFix for less than 2 weeks and application of a non standard ExFix such as Ilzarov type fine wire constructs or hybrid devices. Patients who had fixations applied at different locations were also excluded.

Patients remained in patients for the duration of their treatment and the fixator was on the lower or upper extremity. Most patients were treated for compound fractures. Antibiotics were given intravenously as soon as possible and continued for a total of 48 hours in all cases. Patient ages ranged from 4 to 68 years (median 24 years); there were 16 females and 40 males.

Location: National Referral Hospital in Honiara in the Soloman Islands.


InterventionsPatients acted as their own control. The ExFix constructs normally had 4 pins, 2 proximal and 2 distal. There were 204 pins in 56 patients. The grouped pins were randomised; comparisons were made between proximal and distal groups of pins.

Group 1: daily pin site care: crusts around pins removed, sterile saline irrigation, drying with sterile swabs, sterile dressing pre-soaked in povidone-iodine (Betadine) and applied around pin site and dry dressing wrap.

Group 2: no pin site care

Duration of treatment: median 55 days (range 16 to 158 days).


OutcomesCondition of the soft-tissue interface, stability of the pins, torsional stability of the pins as determined with a torque metre during their removal, osteolysis on pre-removal radiographs, pin site pain


NotesThe authors noted that changes in the soft tissue interface did not necessarily represent an infection but the authors noted that microbiological confirmation of infection was unreliable as this may represent normal skin colonisation.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Random binary number generator"

Allocation concealment (selection bias)Unclear riskNot reported

Blinding of participants and personnel (performance bias)High riskNot reported and unlikely

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"The person assessing the pin sites was blind as to the treatment or not each pin received"

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo patients lost to follow up

Selective reporting (reporting bias)Unclear riskFull details of all prespecified outcomes not reported

Other biasUnclear riskRandomisation per pin site rather than participant - possible unit of analysis errors because data are not independent

Camilo 2005

MethodsTwo-centred parallel group study.
No power calculation.
No of patients randomised = 30.
No of patients analysed = 30.
Source of funding: nil.


ParticipantsInclusion criteria: patients with the Ilizarov external fixation method having their first postoperative return visit.
Exclusion criteria: active infection in the limb being treated.
Patients all had surgery on the femur and/or tibia for combinations of bone lengthening, strengthening, transport, compression and tension.
Patients were all Caucasian with an average age of 31-32 years.
Location: two orthopedic services in Sao Paolo, Brazil.


InterventionsPatients were given instructions for the pin site care and carried it out themselves on a daily basis.
Group 1: 0.9% physiological saline dressing with site covered by gauze dressing;
Group 2: 0.9% physiological saline plus 10% polyvinylpyrrolidone-iodine covered by gauze dressing.
Duration: mean 273 days (range 95-726 days), i.e. length of time the Ilizarov external fixator in place.


OutcomesMeasured at outpatient return visits (not clear when these occurred).
Fever; hyperemia; oedema; purulent secretion (defined as presence of infection); pain; rehospitalisation.


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskRandomisation method not described.

Allocation concealment (selection bias)Low riskConsecutively-numbered, sealed envelopes.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants accounted for.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported

Other biasUnclear riskRandomised groups were not comparable at baseline. There were more males than females.

Cavusoglu 2009

MethodsSingle-centred, parallel group study.
No power calculation.
No of patients randomised = 39.
No of patients analysed = 39.
ITT analysis.
Source of funding: nil reported.


ParticipantsInclusion criteria: not reported, but all participants had circular external fixators applied to the tibia and fibula for their fractures.
Exclusion criteria: not reported.
Mean age of participants: 50 and 47 years, male gender 65% and 53%, periarticular localisation 55% and 16%, open fractures 25% and 47%, follow up 135 and 130 days, according to randomised group (Group 1 and Group 2, respectively)
Location: hospital clinic in Turkey


InterventionsAll patients cleaned their pin sites using sterile gauze impregnated with iodine solution every 3 days for the first 15 days, then randomised to:
Group 1: daily showering and brushing pin sites with soap and toothbrush;
Group 2: daily showering and cleaning pin sites with sterile gauze impregnated with iodine solution.

Duration: up to 150 days, until fixator removal.


OutcomesInfection rates according to modification of a system of Dahl described by Gordon.


NotesInfection rates reported per pin site rather than per patient.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer-generated randomisation sequence.

Allocation concealment (selection bias)Unclear riskNot reported.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo incomplete data.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasUnclear riskInfection rates per pin site rather than per patient means that unit of analysis errors cannot be excluded.

Chan 2009

MethodsSingle-centred, parallel group study.
Power calculation not reported.
No of patients randomised = 60 (but 62 initially entered trial).
No of patients analysed = 60.
Source of funding not reported.


ParticipantsInclusion criteria: not stated.
Exclusion criteria: not stated.
Participants were 46 males and 14 females, mean age 26 years, with Ilizarov external fixators for congenital or developmental deformities or trauma.
Location: hospital in Kuala Lumpur, Malaysia.


InterventionsGroup 1: diluted iodine dressing solution;
Group 2: saline dressing solution.
Duration: 6 months with assessments every 2 weeks.


OutcomesInfection rates (according to 3 grade system developed at the Malaysian hospital)..


NotesOutcome analysed per interface, rather than per patient.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod not reported.

Allocation concealment (selection bias)Unclear riskNot reported.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskTwo reported dropouts prior to randomisation.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasUnclear riskOutcome analysed per interface rather than per patient, so unit of analysis errors cannot be excluded.

Egol 2006

MethodsSingle-centred, parallel group study.
Power calculation for sample size.
No of patients randomised = 118 (120 wrists).
No of patients analysed = 118.
Source of funding: nil.


ParticipantsInclusion criteria: skeletally-mature patients with a displaced, unstable, distal radial fracture that was amenable to the use of external fixators.
Exclusion criteria: not reported.
Location: 2 hospitals (1 academic centre) in New York.


InterventionsAll patients had repair of fractures using a uniplanar external fixator with standard protocol for pin placement and closure. All patients were given non-adherent dressing and antibiotic prophylaxis before and after surgery, and were randomised 1 week postoperatively.
Group 1: daily cleansing with solution of 50% saline solution + 50% hydrogen peroxide;
Group 2: Biopatch dressing over pin sites held in place with sterile gauze (changed weekly);
Group 3: weekly treatment with dry dressing with no pin site care.
In all groups, the fixators were covered with an elastic bandage.
Duration: 6 months with weekly follow-up for assessment of outcomes.


OutcomesMeasured weekly.
Erythema, cellulitis (erythema associated with fever or elevated white blood cell count), drainage, clinical or radiographic evidence of pin loosening, need for antibiotics, need for pin removal before fracture healing due to infection.
Antibiotics were only given if erythema, cellulitis or infection developed.


NotesCompliance not measured.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandom number table of 120 sequential numbers divided into groups of 40.

Allocation concealment (selection bias)Low riskSealed, numbered envelopes in random order.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants appeared to be accounted for.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasUnclear riskBaseline imbalance reported (gender and patients who had open procedures in addition to external fixation). Also, potential unit of analysis errors (120 wrists and 118 patients).

Grant 2005

MethodsSingle-centred, parallel group study.
No power calculation.
No of patients randomised = 20 (131 pins).
No of patients analysed = 18 (116 pins).
Source of funding not stated.


ParticipantsInclusion criteria: > 18 years old; resident within the Health Service District; able to provide written informed consent; admission to hospital with acute injury requiring insertion of skeletal pin or wire.
Exclusion criteria: not reported.
86% of the patients had fractures of the leg, ankle or foot. The majority (> 80%) were male.
Location: Gold Coast Hospital, Australia.


InterventionsAll patients had gauze soaked with 10% povidone-iodine solution at each pin site removed 72 hours after surgery. Pin site care took place 72 hours postoperatively and was mostly performed by the patients.
Group 1: cleansing with normal saline flush + soft white paraffin ointment for 14 days, or until discharge;
Group 2: twice-daily cleansing with normal saline and application of 10% povidone-iodine solution for 14 days or until discharge, then daily care.
All patients continued with their protocol at home until either infection occurred, or pins were removed.
Prophylactic antibiotics were prescribed either during surgery or immediately after for 83% of patients.


OutcomesClinical signs of infection.
Antibiotic use.


NotesEarly antibiotic treatment in some patients may have confounded the results.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskWord 'randomly' used but method unclear.

Allocation concealment (selection bias)Unclear riskNo published evidence of allocation concealment.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
High riskUnclear whether withdrawals came exclusively from one group. Because of small sample size, withdrawals could bias estimates; no ITT analysis.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasHigh riskNo baseline imbalance, but antibiotic use only in some patients, which could confound the estimation of outcomes.

Henry 1996

MethodsSingle-centred, parallel group study.
No power calculation.
No of patients randomised = 30 (120 pin sites - not clear whether the patients or the pin sites were randomised).
No of patients analysed = unclear.
Source of funding: not stated.
Duration 56-244 days.
No inclusion criteria identified.


ParticipantsInclusion criteria: not stated.
Exclusion criteria: not stated.
Patients 11-18 years, having planned lengthening surgery. Pins were external fixators located on the femur or tibia.
Location: hospital in Middlesex, UK.


InterventionsGroup 1: cleansing with 0.9% saline;
Group 2: cleansing with 70% alcohol;
Group 3: no cleansing.
All sites sprayed with povidone iodine and covered with dry gauze.


OutcomesInfection rate.


NotesAfter completion of the trial, neither of the experimental regimens adopted.
Infection rates in pin sites were found to be associated with location on the body.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandom sequence numbers.

Allocation concealment (selection bias)Unclear riskNo published evidence of allocation concealment.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll participants appear to be accounted for.

Selective reporting (reporting bias)High riskSpecification of outcomes for study are unclear.

Other biasUnclear riskNo indication whether baseline comparability between groups. Likely unit of analysis errors because each participant had more than 1 pin site.

Lee 2012

MethodsSingle centre, parallel group study

Power calculation for sample size, but no details reported

No of patients randomised = 38

No of patients analysed = 38

Source of funding: not reported


ParticipantsInclusion criteria: consecutive patients scheduled for limb lengthening or deformity correction using external fixation

Exclusion criteria: patients with multiple underlying medical problems and patients who could not return for regular follow ups in the hospital

Participants had a mean age of 26.3 years; 23/38 (60.5%) were male and 15/38 (39.5%) were female. There were 40 limbs studied among the 38 patients.

Location: Dept of Orthopedic Surgery, University Malaya Medical Centre, Kuala Lumpur, Malaysia.


InterventionsGroup 1: Pin site care with gauze impregnated with 0.2% PHMB (polyhexamethylene biguanide)

Group 2: Plain gauze

Duration: Assessments made at 2, 4, 8 and 12 weeks after surgery


OutcomesPin tract infection (according to Saw 2006 grading system)

Side effects


NotesAnalysis performed according to number of observations made (each pin site had 4 observations) rather than per pin site or per patient so data could not be pooled. Results in the review are reported narratively.

Pin site care was taught to patients or caretakers according to a standard protocol used in the hospital


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomisation by drawing lots

Allocation concealment (selection bias)Unclear riskNot reported

Blinding of participants and personnel (performance bias)Low risk"Patients blinded to the type of gauze they would be using"

Blinding of outcome assessment (detection bias)
All outcomes
Low risk"Pin sites were assessed by two of the authors who were blinded to the type of dressing gauze used"

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo dropouts from study

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported

Other biasHigh riskUnit of analysis errors likely because analysis undertaken according to observations (as each patient had 4 observations at different times, some of the data were not independent)

Patterson 2005

MethodsTwo-centred, parallel group study.
No power calculation.
No of patients randomised = 101.
No of patients analysed = 92 (9 did not complete the data forms).
Source of funding: NAON and AirCast research grants. Kendal Company supplied the Xeroform wraps.


ParticipantsInclusion criteria: not reported.
Exclusion criteria: not reported.
Participants were consecutive patients treated for fractures with skeletal pins or devices. Sample was 2/3 male, 31% were children or teenagers, the most common diagnosis was closed fracture with lower extremity involvement and on average there were six pins/participant.
Location: orthopedic/trauma unit and paediatric unit in hospital in New England and a children's hospital in central USA.


InterventionsGroup 1: half-strength peroxide cleansing, rinse with saline, apply stable gauze/sponge BID;
Group 2: half-strength peroxide cleansing, rinse with saline, apply Xeroform/Xeroflo wraps BID;
Group 3: saline cleansing, apply stable gauze/sponge BID;
Group 4: saline cleansing, apply Xeroform/Xeroflo wraps BID;
Group 5: antibacterial soap and water cleansing, apply stable gauze/sponge BID;
Group 6: antibacterial soap and water cleansing, apply Xeroform/Xeroflo wraps BID;
Group 7: no cleansing, apply gauze/sponges;
Duration: 6-week follow-up visit, or pin removal.


OutcomesInfection rate (patients and pins).


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod not described.

Allocation concealment (selection bias)Unclear riskNo published evidence of allocation concealment.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
High riskDropouts reported, but not clear which group(s) they belonged to.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasHigh riskBaseline comparability between groups not reported. Unit of analysis errors likely as each participant had more than 1 pin site.

W-Dahl 2003

MethodsSingle-centred, parallel group study.
Power calculation for sample size.
No of patients randomised = 50.
No of patients analysed = 50.
Source of funding: not stated.


ParticipantsInclusion criteria: not stated.
Exclusion criteria: not stated.
All patients had operation for gonarthrosis by hemicallotasis technique.
Location: hospital in Sweden.


InterventionsGroup 1: daily pin site care;
Group 2: weekly pin site care.
Pin site care for all patients was cleansing with 0.9% NaCl solution, then application of sterile compress fixed with soft dressing. All patients had prophylactic antibiotics pre surgery and for 14 days post surgery.
Follow-up once a week after discharge from hospital.


OutcomesInfection rate (defined by Checketts-Otterburns), cultures (negative/positive), use of antibiotics (days), pain (VAS scale).


NotesAuthor answered request for clarification of outcomes.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskTable of random numbers.

Allocation concealment (selection bias)Unclear riskNo published evidence of allocation concealment.

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskBlinding highly unlikely as different care regimen

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing participants.

Selective reporting (reporting bias)Low riskAll prespecified outcomes reported.

Other biasUnclear riskBaseline comparability between groups, but likely unit of analysis errors because each participant had more than 1 pin site.

Yuenyongviwat 2011

MethodsSingle centre, parallel group study

Power calculation for sample size (required 40 pins per group to find a difference of 30% in pin site infection rates with alpha 0.05 and beta 0.2)

No of patients randomised = 30

No of patients analysed = 30

Source of funding: not reported


ParticipantsInclusion criteria: patients with open tibial fracture requiring external fixation

Exclusion criteria: patients with immunocompromised condition (diabetes and HIV) and malignancy

Emergency debridement and irrigation was followed by use of an AO fixator.

Participants had a mean age of 29 and 30 years in the intervention and control groups respectively and half were male. Cause of injury was motorcycle accident in 23 cases and was a comminuted fracture pattern in 24 cases

Location: Department of .Orthopaedic Surgery and Physical Medicine at Prince Songkla University, Hat Yai, Songkla, Thailand


InterventionsGroup 1: daily pin site dressing with normal saline and application of 0.5ml of 1% silver sulfadiazine

Group 2: daily dry dressing with optional removal of dry scale by patients

Treatments were applied on day 3 after surgery. Mean duration of parenteral antibiotics was 9.9 and 12 days.

Duration: mean follow up was 106.3 and 108.6 days in the intervention and control groups respectively. Assessments were made every month


OutcomesPin site infection rate (according to Checketts-Otterburns classification)


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskMethod not reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding of participants and personnel (performance bias)High riskBlinding highly unlikely as different care regimen

Blinding of outcome assessment (detection bias)
All outcomes
High riskAssessors not blinded

Incomplete outcome data (attrition bias)
All outcomes
Low riskNo dropouts

Selective reporting (reporting bias)Unclear riskAll prespecified outcomes reported but not clear when assessments were made

Other biasUnclear riskNot possible to determine whether groups were comparable at baseline because group demographics not clearly reported.

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Bednar 1996This study focused on the use of external fixator in the management of back pain, it had a specific pin site care protocol but did not include this as part of the research.

Bhattacharya 2006Participants allocated consecutively, and not randomised.

Botte 1989This prospective clinical trial considered the need for stab wound incision prior to the placement of halo skeletal fixator pins. It did not meet any inclusion criteria.

Davies 2005Allocation to groups was not random.

Hutchinson 2000This RCT considered the advantages of pre-drilling before the insertion of external fixation pins in treating distal radius fractures. A pin care regime was prescribed but not considered as part of the research.

Kapoor 2000This RCT study considered three different methods of treating distal radius fractures. Their findings recommended that displaced severely-comminuted intra-articular fractures should be treated with an external fixator. They mentioned pin site care but did not consider it in the trial.

Lagerquist 2012Systematic review

Masse 2000This RCT compared silver coated and standard stainless steel screws used with an external fixator. A pin site regimen was included but not considered as part of the trial.

Moroni 2001aThis RCT considered the application of external fixation for wrist fractures in osteoporotic women, specifically improvement in the bone-pin interface - and thus fixation of pins - and compared hydroxyapatite-coated pins with standard tapered pins. There was no information about pin site care, but considerable detail on pin site infection.

Moroni 2001bThis RCT compared biomechanical and clinical properties of standard tapered pins and hydroxyapatite-coated pins in patients with femoral or tibial fractures treated with external fixation. A pin site cleaning regime was stated but was not discussed as part of the research.

Sproles 1985A CCT where patients were allocated by location in hospital, therefore, experimental and control groups were systematically different for reasons other than pin site care.

Turcic 1998This study considered the outcome of converting original external fixation treatment to screw fixation compared with maintaining the external fixation for the duration of treatment of fractures due to war wounds. A wound care protocol was included, but not discussed, in the research.

Vossinakis 2002This RCT compared the effectiveness of an external fixator with a sliding hip screw in the treatment of pertrochanteric femoral fracture. Entry points for pins were prevented from tenting, and a daily dressing regimen was carried out, but not considered as part of the trial.

W-Dahl 2004The participants in this study were allocated consecutively, and were not randomised to groups.

 
Comparison 1. Cleansing compared with no cleansing

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Infection rates3295Risk Ratio (M-H, Random, 95% CI)2.30 [0.63, 8.33]

 
Comparison 2. Sterile antiseptic cleansing solution compared with sterile non-antiseptic cleansing

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Infection rate3920Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.82, 1.39]

 
Comparison 3. Sterile cleansing method compared with non-sterile cleansing

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Minor infection1610Risk Ratio (M-H, Fixed, 95% CI)0.86 [0.73, 1.02]

 2 Major infection2691Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.49, 1.30]

 
Comparison 4. Dressing compared with no dressing

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Successful condition of soft tissue interface1204Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.70, 1.48]

 
Comparison 5. Comparisons between different types of dressings

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Antibiotics needed1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 2 Clinical signs of infection1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 3 Superficial infection1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 4 Infection rate xeroform vs gauze or sponge1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 5 Any infection sulfadiazine vs dry dressing1Risk Ratio (M-H, Fixed, 95% CI)Totals not selected

 
Table 1. Summary data table

Study IDNParticipantsInterventionsInfection rateDefinition of infectionDurationAntibiotic prophylaxis

Camilo 200530Ilizarov external fixators; no infection.(1) Daily: shower, cleanse with saline, gauze dressing soaked with polyvinylpyrrolidone-iodine;

(2) Daily: shower, cleanse with saline; dry dressing
(1) 66.7%; (2) 46.7%; NSDefined as purulent secretion.Follow-up for time external fixator in place, mean = 273 days (95-726 days)Not reported.

Egol 2006118Adults; displaced, unstable, distal radial fracture(1) Daily: cleansing with 1/2 saline/1/2 hydrogen peroxide;

(2) Biopatch dressing changed weekly;

(3) No pin site care; dry dressing changed weekly.
(1) 22.5%; (2) 5%;
(3) 2.5%.
Defined as requiring antibiotics.Follow-up for mean of 5.9 weeks (4-8 weeks). External fixator removed at 6 weeks and patients followed-up for 6 months.Standard prophylaxis given to all patients: before and immediately after surgery.

Grant 20052018+ years(1) Daily: saline cleansing, soft white paraffin ointment;

(2) Twice daily: saline cleansing, 10% povidone-iodine.
(1) 34.1%; (2) 18.1% (pin sites, not participants).Defined as redness, induration, haemo-serous ooze and pain.Follow-up until clinical infection or removal of pin.Intraoperative or postoperative period: 83%.

Henry 199630Aged 11-18 years(1) Daily: cleansing with 0.9% sodium chloride;

(2) Daily: cleansing with 70% alcohol;

(3) Daily: no cleansing.

All groups had pin site care consisting of sterile gauze covering, ease of crusting; massage, dry povidone-iodine spray.
(1) 25%;

(2) 17.5%;

(3) 7.5%. NS
Defined as pain, redness and swelling and when a significant number of pathogenic bacteria cultured.Pin in place for mean of 150 days (56-244 days)Not reported.

Patterson 2005101All ages (3-80 years); 65% male; no inclusion criteria reported.(1) Twice daily: 1/2 strength peroxide, rinse with saline, apply stable gauze/sponge;

(2) Same cleansing; apply Xeroform/Xeroflo dressing;

(3) Twice daily; saline cleansing, apply stable gauze/sponge;

(4) Same cleansing; apply Xeroform/Xeroflo dressing;

(5) Twice daily: antibacterial soap and water cleansing, apply stable gauze/sponge;

(6) Same cleansing; apply Xeroform/Xeroflo;

(7) No cleansing, apply gauze/sponge (change only if wet/soiled).
(1) 46%;

(2) 9%;

(3) 33%;

(4) 27%;

(5) 39%;

(6) 50%;

(7) 36%. NS
Rating scale that measured redness, swelling, discomfort, tenting, loosening of pins, crusting and drainage. Stage II infection defined as score >3 and requiring treatment with antibiotics; stage III infection defined as score > 7, treatment with IV antibiotics and/or removal of pin.6-week follow-up.Not reported.

W-Dahl 200350Surgery for gonarthrosis by hemicallotosis technique; mean age 54 years; inclusion criteria not reported.(1) Daily: pin site care: cleansing with 0.9% sodium chloride, sterile compress fixed with soft dressing;

(2) Weekly; same pin site care protocol.
Grade 1:
(1) 7.4%;

(2) 12%.

Grade II:
(1) 2.8%;

(2) 0%. NS
Checketts-Otterburns classification of pin site infections (Checketts 1999).10-week follow-upProphylaxis with IV antibiotics during surgery and 2 doses within 24 hours post operatively followed by 14 days of oral prophylaxis.

Cavusoglu 200939Mean age 50 and 47 years in two groups; 65% and 53% male; no inclusion criteria reported.(1) Daily showering; cleaning crusts with sterile gauze with iodine solution;

(2) Daily showering; brushing pin sites with soap and water.
Minor infection:

(1) 44%;

(2) 51%.

Major infection:

(1) 4%;

(2) 4%.
Minor infection: Grade 1 and 2 on a modification of the Dahl system by Gordon - not requiring any extra pin site care. Major infection: Grade 3 and above - requiring treatment with oral antibiotics.Pin site inspection on 5th, 10th, 15th, 30th, 45th, 60th, 75th, 90th, 120th and 150th days of follow-up until fixator removal.Prophylaxis for closed fractures: preoperative and postoperative single dose intravenous cephazolin sodium (1 g); open fractures: preoperative 1 day and postoperative 4 day intravenous cephazolin sodium (1 g 3 X day), intravenous gentamicin (160 mg/day) and oral ornidazole (500 mg 2 X day).

Chan 200962Distraction osteogenesis using Ilizarov fixators; age ranged from 1-76 years; 37% < 14 years; no inclusion criteria reported.(1) Daily povidone-iodine cleansing solution;

(2) Saline cleansing solution.
Any grade of Infection:

(1) 19%;

(2) 17%.
Three point grading of infections by system developed in Department of Orthopaedic Surgery in University of Malaya Medical Centre.Every 2 weeks for 6 months.Antibiotics only prescribed for patients with grade 2 or 3 infections.

Lee 201238Patients scheduled for limb lengthening or deformity correction; mean age 26.3 years (range 5 to 68 years) with 61% male(1) PHMB impregnated gauze

(2) Plain gauze
Any grade of infection

(1) 11 of 1068 observations

(2) 39 of 864 observations
Saw classification of pin site infections (Saw 2006).Blinded assessments made at 2, 4, 8 and 12 weeks after surgery.Not reported

Yuenyongviwat 201130Patients with open tibial fracture; mean age 29 and 30 years mostly male (83%)(1) daily dressing with normal saline and 0.5ml of 1% silver sulfadiazine

(2) daily dry dressing
Any infection

(1) 7/15 (47%)

(2) 6/15 (40%)
Classification system developed by Checketts 1999.Assessments made by independent observer monthly for up to 3 months (106 and 108 days follow up).Mean duration of parenteral antibiotic as 9.9 and 12 days in each group, respectively.

Camathias 201256Patients with open fractures or other diagnoses in the upper or lower extremities; no further details(1) daily pin site care (removal of crusts, sterile saline irrigation, drying with sterile swabs, sterile dressing pre-soaked with povidone-iodine, dry dressing wrap

(2) no pin site care
Condition of the soft tissue interface (method not described)Soft tissue interface classified as dry, secretions or granulation. Dry pin sites regarded as a successful outcome and secretions or granulation regarded as a failure.Blinded assessments made at a median of 55 days (range 16 to 355 days)All patients had intravenous antibiotics as soon as possible for a total of 48 hours.

 PHMB: polyhexamethylene biguanide
NS: not significant