Early vs. Late Tourniquet Release and Phlebotomy-Induced Hemolysis in the Emergency Department: TOURNI-ED

BACKGROUND AND RATIONALE:

Hemolysis is recognized as the leading preanalytical source of error in clinical laboratories, with emergency department settings reporting particularly high rates (range: 12-26%) due to specimen collection under time pressure, use of small-bore intravenous catheters, and challenging venous access. Hemolyzed specimens affect measurements of potassium, lactate dehydrogenase, bilirubin, and other analytes, resulting in clinical misinterpretation, repeated laboratory requests, and procedural delays in a time-sensitive environment.

Venous stasis created by tourniquet application is a physiological perturbation that promotes erythrocyte deformation and lysis through hemoconcentration and shear stress. Clinical guidelines (Clinical and Laboratory Standards Institute, CLSI H03-A6; Turkish Biochemistry Society Phlebotomy Guidelines 2015) recommend that the tourniquet should be released as soon as blood flow is established, ideally within 60 seconds, and not later than the completion of the first collection tube. However, in routine emergency nursing practice, the tourniquet is frequently maintained throughout the entire multi-tube collection sequence to maintain venous distension and reduce the likelihood of failed collection attempts.

This pragmatic trial directly tests whether adherence to guideline-recommended early tourniquet release translates into a clinically meaningful reduction in hemolysis rates compared to the commonly observed late release practice, in a controlled setting using standardized sphygmomanometer-applied venous stasis at 60 mmHg.

INTERVENTION DELIVERY:

Both interventions use a standard aneroid sphygmomanometer (not a conventional tourniquet rubber strap) inflated to 60 mmHg to standardize venous stasis pressure across all participants. Applying stasis with a sphygmomanometer holds the pressure fixed and measurable, so that pressure does not vary between operators and the arms differ only in the timing of cuff release.

Tube collection order (identical for both groups, per CLSI H03-A6):

Sodium citrate (blue cap, 2.7 mL) - coagulation studies SST/gel separator (yellow cap, 5 mL) - serum biochemistry [PRIMARY OUTCOME TUBE] K2-EDTA (purple cap, 3 mL) - complete blood count

Additional tubes (for example a second biochemistry tube or a troponin tube) are drawn when ordered by the treating physician as part of routine clinical care; these additional tubes are not part of the study protocol. The total number of tubes per participant therefore ranges from 3 to 5.

Group A - Early Release Protocol: As soon as blood flow into the first tube (sodium citrate) is observed, the sphygmomanometer is deflated to 0 mmHg and removed. Subsequent tubes are collected without venous stasis.

Group B - Late Release Protocol: The sphygmomanometer remains inflated at 60 mmHg throughout the entire collection sequence. It is deflated and removed only after the last collection tube completes filling.

RANDOMIZATION AND ALLOCATION CONCEALMENT:

The allocation sequence was generated by the principal investigator before the start of enrollment, using Research Randomizer (randomizer.org), as a restricted list of 792 allocations in a 1:1 ratio without stratification. The block length is not reported, because the block structure of the generated list was not preserved in the sequence as implemented. The complete sequence was fixed before the first participant was enrolled; no sealed envelopes were used. The sequence was embedded in the electronic data collection application used at the bedside and was not accessible to the enrolling nurse; the application released a participant's allocation only after eligibility had been confirmed and the participant had been enrolled, so allocation concealment was maintained up to the point of enrollment. The realized allocation was 396 participants per group, and the largest cumulative imbalance between the groups at any point during enrollment was 4 participants.

DIFFICULTY ASSESSMENT - A-DIVA Scale: All participants are assessed for venous access difficulty using the modified A-DIVA (Adult Difficult Intravenous Access) Scale (van Loon et al., J Clin Med 2019;8(2):144) prior to blood collection. The modified A-DIVA scale assigns 0-1 points for each of five items: (1) history of difficult venous access, (2) no palpable vein, (3) no visible vein, (4) vein diameter <3 mm, (5) difficult venous access anticipated by the operator. Total score ≥2 is classified as high-difficulty. The A-DIVA score is recorded as a covariate in statistical analysis.

HEMOLYSIS ASSESSMENT:

The Hemolysis Index (HI) is measured on all SST/yellow cap tube specimens by the clinical chemistry analyzer (DxC series; Beckman Coulter, Brea, CA), which reports each serum specimen either as negative for hemolysis or as an integer hemolysis index. The laboratory maps these values onto a six-category ordinal scale corresponding to free hemoglobin concentrations: (-) <50 mg/dL, (1+) 50-99 mg/dL, (2+) 100-199 mg/dL, (3+) 200-299 mg/dL, (4+) 300-500 mg/dL, (5+) >500 mg/dL. Laboratory technicians performing HI analysis are blinded to group assignment throughout the study.

BLINDING:

Participants and the nurse performing venipuncture could not be blinded: the consent process described both release timings, and the timing of cuff release is visible at the bedside. The outcome assessor (the laboratory technician recording the hemolysis index) was blinded to allocation, as was the analyst until the analysis plan, which was fixed before enrollment began, was finalized. Each tube carried its own laboratory information system barcode; the serum separator gel tube was identified by that barcode alone, the allocated group appeared neither on the request nor on the tube, and the hemolysis index recorded against that barcode was the value transcribed into the data set.

PATIENT AND PUBLIC INVOLVEMENT (CONSORT 2025 Item 8):

Patients and the public were not formally involved in the design of this trial due to the pragmatic, low-risk, procedure-modification nature of the study. Findings will be disseminated through open-access publication and presented to nursing and laboratory staff at the participating institution.

DATA SHARING (CONSORT 2025 Item 4):

De-identified individual participant data and the data dictionary will be made available to qualified researchers upon reasonable written request to the principal investigator, following publication of the primary manuscript and approval by the institutional ethics committee, in accordance with applicable data protection regulations.

Inclusion criteria

• Age 18 years or older
• Presenting to the emergency department and triaged as green (semi-urgent) or yellow (urgent) category according to the Emergency Severity Index (ESI) or equivalent institutional triage system
• Blood collection (venipuncture) indicated as part of routine clinical care by the attending emergency physician
• Ability to provide written informed consent
• Accessible peripheral vein suitable for standard venipuncture, most often in the antecubital fossa but at the operator's discretion an alternative peripheral site such as the dorsum of the hand (not requiring intravenous catheter placement for blood collection)

Exclusion criteria

• Blood collection performed via an existing intravenous catheter or central venous access device
• Critical illness or hemodynamic instability, including triage category red (resuscitation) at the time of blood collection
• Anticoagulant or thrombolytic therapy, or any condition in which collection of a coagulation tube was contraindicated, including a known or suspected coagulation disorder (e.g., hemophilia, thrombocytopenia with platelet count <50,000/uL)
• Known hemolytic disease, a history of active hemolysis, or another hematological condition associated with baseline elevated hemolysis
• Red blood cell transfusion within the preceding 24 hours
• Pregnancy of any gestational age
• Absence of a peripheral vein suitable for standard venipuncture
• Local pathology preventing collection from the antecubital fossa: infection, hematoma, burn, active upper extremity injury, lymphedema, an arteriovenous fistula, or the side of a previous mastectomy
• Only a single collection tube ordered (the intervention requires at least two tubes, because it is defined by the timing of cuff release relative to tube filling)
• Inability to give informed consent, or declining to participate
• Previously enrolled in this study (re-enrollment not permitted)