Study design

The trial is designed as a randomised, investigator and outcome assessor blinded single-centre trial with two parallel groups, and a primary end point of total range of movement of the affected finger after 3 months of surgery.

Setting

The recruitment of the patients will take place in the emergency department of the Tampere University Hospital, Tampere, Finland. The hospital is responsible for providing treatment of acute hand injuries and infections to 900 000 inhabitants.

Participants

Inclusion criteria

• Clinical diagnosis of purulent flexor tenosynovitis with all four positive Kanavel signs2

  • Symmetrical swelling of the entire digit

  • Exquisite tenderness along the course of the tendon sheath

  • Semiflexed posture of the digit

  • Pain with attempted passive extension of the digit

• Age over 18 years

• Patient's willingness to participate in the study

Exclusion criteria

• High pressure, foreign body or chemical injuries that require open debridement

• Being a prisoner or military serviceman, or being mentally retarded or having other factors that may affect decision-making.

Interventions

After clinical examination, laboratory tests and filling in the baseline Quick Disabilities of the Arm, Shoulder and Hand Score (QuickDASH) form,14 tendon sheath irrigation is performed in the operating room identically in both study arms until randomisation. The procedure for intraoperative irrigation of the infected flexor tendon sheath is a modification of a guideline described by Gutowski et al12 (figure 1). The flexor tendon sheath is opened proximal to the A1 pulley of the affected finger. Bacterial cultures are collected and the appearance of the exudate is noted. An 18-gauge angiocatheter is inserted percutaneously into the wound, approximately 1 cm proximal to the incision. The tip of the catheter is placed within the sheath under the A1 pulley and the catheter is secured to the skin, using a suture. A counter incision is made midaxially at the level of the A4 pulley. A midaxial incision is favoured in order to avoid inconvenient scarring on the palmar surface of the finger. The tendon sheath is irrigated with 50 mL of physiological saline through a proximal catheter. The irrigation is continued until the output is clear. A small rubber drain is placed in both incisions to keep them open. If the thumb is involved, the catheter is placed in the flexor pollicis longus sheath distal to the carpal tunnel. If needed, a separate incision is made radial to the flexor carpi radialis tendon to drain the most proximal part of the tendon sheath. Depending on the group allocation after randomisation, the catheter is thereafter removed or retained. A normal hand dressing is applied.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Schematic presentation of catheter irrigation. Incision (red) for placement of catheter tip (black) underneath A1 pulley (blue) and midaxial counter-incision (red) for outflow.

For the patients of the intraoperative and postoperative irrigation group, the postoperative irrigation is performed by specially trained nurses in the ward, using 20 mL saline three times a day for 3 days. On day 3, the tip of the removed angiocatheter is sent for bacterial culture. Hand therapy is initiated as early as possible in the ward. Antibiotic treatment is initiated in the operating room after the bacterial samples are collected. The primary antibiotic treatment is cefuroxime 1.5 g three times a day. The secondary choice (in case of allergy) is clindamycin. After discharge, the peroral antibiotic (primarily cephalexin) is continued for 10 days. The patients in both study arms receive identical written instructions for postoperative mobilisation.

Outcome measures

The patients have two follow-up visits at the outpatient clinic 4 weeks and 3 months postoperatively.

The primary outcome measurement is

• Total active range of movement of the most affected finger 3 months postoperatively

The secondary outcome variable is

• Need for reoperation during the first three postoperative months

Other outcome variables are

• QuickDASH score14 (4 weeks and 3 months postoperatively)

• Pain at rest (visual analogue scale; 4 weeks and 3 months postoperatively).

Allocation and blinding

The patients are stratified in four groups depending on the purulence of the exudate (clear vs murky or purulent) and age over 43 years or the presence of diabetes mellitus, peripheral vascular disease, or renal failure (yes vs no), which have been shown to be associated with poor outcome after purulent tenosynovitis.15 A block randomisation to two study arms (intraoperative irrigation only or intraoperative and postoperative irrigation) is carried out within these four groups in order to ensure even allocation. Only the statistician who carried out the randomisation is aware of the size of the block. The assignments are enclosed in opaque, sealed envelopes that are sequentially numbered for each stratification group.

The patient cannot be blinded. The operating surgeon is blinded until the randomisation. The randomisation is delayed to take place just after the intraoperative irrigation in order to ensure the longest possible blinding of the surgeon. The staff in the ward cannot be blinded. The investigator in the outpatient clinic is blinded, because he/she has not participated in the treatment (neither in the operation nor in the ward).

Data collection and monitoring

The patient recruitment and treatment is performed by senior hand surgeons or hand surgery residents of the Tampere University Hospital, Tampere, Finland. The collection of narrative and objective data is saved in a password-protected database. Only members of the Data Monitoring Board have access to the database during the study period. All adverse events (AE) will be documented in detail, and will be reported to the Data Monitoring Board. The principal investigator will report the AE within 24 h after it becomes known. The investigators are responsible for making the final decision to terminate the trial. Participants who suffer an AE will be given adequate medical treatment and will be entitled to seek compensation from the Finnish Patient Insurance Centre.

Sample size

A total of 48 patients (24 patients/group) are needed for the study. This is based on the power calculation: p=0.05, p=0.8, group difference 20% (in the total range of movement at 3 months postoperatively) and relative SD 20%, drop out 25%.

Patient timeline

Figure 2 shows the patient timeline.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

The schedule of enrolment, interventions and assessments demonstrated in the SPIRIT figure.

Statistical analysis

All analyses will be performed according to the intention to treat principle. Analysis of variance is used for all numeric variables (range on movement, QuickDASH-score, pain score). The χ² test is used in the comparison of incidences of reoperation. For all tests, we will use two-sided p values with a p<0.05 level of significance.