SPOT Grade: A New Method for Reproducibly Quantifying Surgical Wound Bleeding (PDF)

CE: A.U.; SPINE154984; Total nos of Pages: 8; SPINE154984 SPINE Volume 42, Number 00, pp 000–000...

CE: A.U.; SPINE154984; Total nos of Pages: 8; SPINE154984 SPINE Volume 42, Number 00, pp 000–000 ß 2017 Wolters Kluwer Health, Inc. All rights reserved. SURGERY The SPOT GRADE A New Method for Reproducibly Quantifying Surgical Wound Bleeding William D. Spotnitz, MD, MBA,,y Dirk Zielske, MD,y Valerie Centis, PhD, Rachel Hoffman, BSE,z Daniel L. Gillen, PhD,§ Catherine Wittmann, PhD,y Vincent Guyot, MSE,y Doris M. Campos, PhD,y Patricia Forest, PhD,y Alan Pearson, PhD,{ and Paul C. McAfee, MD, MBA jj 0.9777, 0.9945), physicians correctly identified true hemostasis Study Design. Benchtop model with prospective surgeon video (score ¼ 0). Based upon these data the probability of a physician testing. rating a bleed incorrectly as hemostasis (score ¼ 0) is estimated Objective. To create a surface bleeding severity scale, the to be 1.51% (95% CI: 0.0061, 0.0363). SPOT GRADE (SG), for quantitative assessment of target bleed- Conclusion. This SG is reproducible and reliable providing a ing site (TBS) blood loss. This is of particular interest for spinal basis for educating surgeons on TBS blood loss. It appears to be surgery due to epidural bleeding and an inability to use a new standard for evaluating wound blood loss. diathermy and radiofrequency cautery close to nerve roots. Key words: benchtop model, SPOT GRADE, surface bleeding Summary of Background Data. A novel apparatus perfusable severity scale, target bleeding site, videos. at known flow rates and simulating different sized wounds was Level of Evidence: 2 used to create movies to educate surgeons on specific degrees of Spine 2017;42:xxx-xxx bleeding. Methods. Training (36) and testing (108) videos were created using a benchtop apparatus employing different bleeding seve- rities based on the six-level SG (none, minimal, mild, moderate, H emorrhage and its associated consequences, such severe, and extreme) and TBS sizes (1, 10, and 50 cm2). as transfusion, have gained increased scrutiny over Fourteen surgeons in four specialties (cardiothoracic, abdominal, the past two decades. This is true for medicine in spine, and orthopedic lower extremity) were trained and tested general,1 acute medicine and trauma, and elective surgery.2– to evaluate SG characteristics including inter-rater and intrarater 6 A recent study from a major US surgical center found that reliability. 55% of all perioperative complications and 27% of all Results. The interclass correlation coefficient was estimated to deaths were hemorrhage or thrombosis related.6 This is be 0.89840 (95% confidence interval [CI]: 0.85771, 1), whereas particularly true in spinal surgery with the advent of multi- the intraclass correlation coefficient was estimated to be level spinal osteotomy surgery (Smith-Peterson osteotomy, 0.93673 (95% CI: 0.89603, 1). In 98% of cases (95% CI: pedicle subtraction osteotomy, and vertebral column resec- 0.9736, 0.9927), surgeons correctly identified eligible bleeds for tion) and combined anterior-posterior surgery in the mod- a future clinical trial (scores ¼ 1, 2, or 3) and in 91% of cases ern treatment of adult spinal deformities. (95% CI: 0.8895, 0.9344), surgeons correctly identified none- Despite best efforts, blood transfusions are still associ- ligible bleeds (scores ¼ 4 or 5). In 98.6% of cases (95% CI: ated with the risk of viral transmission, immune modulation leading to general infection, local wound healing compli- From the Department of Surgery, University of Virginia, Charlottesville, cations, organ failure, and even death.5–12 In addition, VA; yBiom’up, SA, Saint-Priest, France; zNorth American Science Associ- ates, Inc, Minneapolis, MN; §Department of Statistics, University of blood transfusions, themselves, are a costly treatment.13,14 California, Irvine, CA; {Origin Product Design, Ltd, Broughton, Cambs, Most recently the question of complications and death UK; and jjScoliosis and Spine Center, University of Maryland, Towson, MD. associated with the use of aging packed red blood cells Acknowledgment date: May 4, 2017. First revision date: September 3, 2017. emphasized that blood transfusions are a limited resource Acceptance date: September 29, 2017. and that reasonable steps should be taken to minimize The manuscript submitted does not contain information about medical device(s)/drug(s). their use.15 Biom’up (Lyon, France) funds were received in support of this work. However, it is still a challenge in research as well as in Relevant financial activities outside the submitted work: consultancy, clinical practice to estimate blood loss directly. State-of-the- employment, travel/accommodations/meeting expenses. art studies in the setting of major trauma took transfusion Address correspondence and reprint requests to William D. Spotnitz, MD, requirements as a ‘‘best proxy’’ for blood loss.16 Similar Department of Surgery, University of Virginia, 2738 SW 92nd Dr, Gaines- indirect attempts are used in elective settings where blood ville, FL 32608; E-mail: [email protected] loss is calculated.17 These indirect attempts are useless when DOI: 10.1097/BRS.0000000000002447 it comes to evaluating local hemorrhage as the blood loss in Spine www.spinejournal.com 1 Copyright ß 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. CE: A.U.; SPINE154984; Total nos of Pages: 8; SPINE154984 SURGERY The SPOT GRADE  Spotnitz et al these settings is usually not affecting the overall physiology. To measure local blood loss, an early surface bleeding severity scale (SBSS) was used in orthopedic surgery, but never validated.6 A review suggested that estimation of wound bleeding by observation might never be a reliable method of determining the severity of wound bleeding.18 More recently, several investigators have used qualitative scales to estimate wound bleeding.19–21 In one study, video segments based on qualitative differences were used to educate surgeons on bleeding scale levels.22 In the most recently published scale, video segments based on sponge weight determinations to quantify the amounts of wound bleeding were used to develop and validate wound bleeding severity determinations.23 However, no previous publica- tions have reported on quantitative video bleeding severity scales based on predetermined known controlled levels of blood flow with as many as six levels of severity. In order better quantify the severity of local bleeding and help describe the efficacy of local hemostatic agents, an SBSS, the SPOT GRADE, was developed and validated. Spinal surgery is an especially critical specialty currently requiring an objective, quantifiable rate of blood loss due to the significant amounts of hemorrhage associated with spinal deformity surgery. To properly gauge the effective- ness of recent innovations in hemostasis in spinal surgery it is more important than ever to have a consistent bleeding scale with a level 3 or below degree (minimal, mild, and Figure 1. Benchtop device for simulating surgical wound bleeding moderate) of hemorrhage. The following innovations in on a plexiglass plate at known flow rates. other specialties have not been fully adopted in spinal surgery for this reason—argon beam coagulators, tranexa- mic acid, hypertonic saline-augmented radiofrequency ablation, and systemic hypotension. As neural tissue is 2 peristaltic pump, SMI-Labhut Limited, Churcham, Glou- particularly vulnerable to diathermy, cerebrospinal fluid cestershire, UK) using tubing (Masterflex LS14 and LS25, is highly conductive of electric current, and there is difficulty Cole Palmer Instrument Company, Ltd, London, UK) at in visualizing epidural bleeding after dural leakage, it is known pump flow rates (0.1–500 mL/min) and measured important to have a bleeding scale appropriate to the loca- pressures (Gems Pressure Transducer 2200RAB1001A3UA, tion and velocity of bleeding commonly faced by spinal Gems Sensors, Ltd, Basingstoke, Hampshire, UK). deformity surgeons. A validated quantitative bleeding scale The apparatus was used to prepare a variety of bleeding will permit hemostatic agents and innovations to limit blood severity videos to simulate surgeons’ procedures in the loss to be precisely evaluated in clinical trials enrolling spinal operating room (See Video, Supplemental Digital Content deformity patients. 1, http://links.lww.com/BRS/B310 which demonstrates a single 10 s training video). MATERIALS AND METHODS The apparatus inflow rate was compared to the outflow rate (sponge weight changeover 5 s) to confirm appropriate Benchtop Apparatus function of the model. A unique new benchtop apparatus (Biom’up, Lyon, France and Origin Product Design, Cambridgeshire, UK) was devel- Video Preparation oped (Figure 1—picture of benchtop apparatus) using a Videos were prepared using constant camera perspective, series of three (1, 10, and 50 cm2) interchangeable perfo- lighting, and magnification (Figure 3—photo of entire sys- rated (0.5 mm) plexiglass plates (Perspex cast Acrylic 3 mm tem) to create bleeding matching the SPOT GRADE scale sheet, Red 440, Perspex, Leeds, Yorkshire, UK) simulating (Figure 4—scale values). The scale values were based on different wound sizes (Figure 2—close up view of three several elements including: visual appearance, expected plates) capable of being perfused with simulated blood intervention(s), and maximum expected American College (Synthetic Blood, Jeulen, Evreux Cedex, France) mixed with of Surgeons – Advanced Trauma Life Support (ACS-ATLS) surfactant (Fairy Original Washing Liquid, Proctor and shock risk class.24,25 Gamble, Weybridge, Surrey, UK) (mixture was 4.76% A surgeon defined the classification of sequential flow surfactant) by a calibrated peristaltic pump (Labhut Allegro rates as falling into the various scale categories and the 2 www.spinejournal.com Month 2017 Copyright ß 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. CE: A.U.; SPINE154984; Total nos of Pages: 8; SPINE154984 SURGERY The SPOT GRADE  Spotnitz et al SPOT GRADETM (SBSS – Surface Bleeding Severity Score) SPOT GRADETM 0 1 2 3 4 5 Severe; not Extreme; Verbal immediately immediately None Minimal Mild Moderate Descriptor life- life- threatening threatening Visual Dry Oozing Pooling Flowing Streaming Gushing Descriptor Manual Manual Manual Manual Manual pressure, pressure, pressure, pressure, pressure, Expected cautery, cautery, cautery, cautery, None cautery, Intervention(s) suture, suture, suture, suture, adjuvant adjuvant adjuvant staples, staples, hemostat(s) hemostat(s) hemostat(s) tissue repair tissue repair Maximum Expected ACS-ATLS1 1 1 1 2 3 4 Shock Risk Class 1 ACS-ATLS Shock Risk Class: 1 – involves up to 15% of blood volume; typically no change in vital signs and fluid resuscitation is not usually necessary. Class 2 – involves 15-30% of total blood volume; patient is often tachycardic with a narrowing of the difference between the systolic and diastolic blood pressures; the body attempts to compensate with peripheral vasoconstriction; skin may start to look pale and be cool to the touch; volume resuscitation with crystalloids is all that is typically required; blood transfusion is not typically required. Class 3 – involved loss of 30-40% of circulating blood volume; patient’s blood pressure drops; heart rate increases, peripheral hypoperfusion worsens; fluid resuscitation with crystalloid and blood transfusion are usually necessary. Class 4 – involves loss of > 40% of circulating blood volume; the limit of the body’s compensation is reached and aggressive resuscitation is required to prevent death. Figure 4. SPOT GRADE bleeding severity scale including definitions and factors influencing scale boundaries including: visual appear- ance, expected intervention(s), and maximum expected ACS-ATLS shock risk class. boundaries for the scale levels were determined (Table 1— flow rates for scale levels). The videos were used to educate (36 videos) and test (108 videos) an invited advisory board of 14 surgeons (United States n ¼ 13; Germany n ¼ 1) from four specialties (cardio- thoracic , abdominal , spine , and lower extremity orthopedic ). The average number of years of clinical experience in each specific surgical indication was 15.6. Training Figure 2. Static appearance of simulated bleeding on different size Thirty-six videos in increasing order of bleeding severity wounds (1, 10, and 50 cm2) with different severities (0, 3, and 4). using the three different plexiglass wound plates (1, 10, and 50 cm2) with two videos (near the upper and lower scale level flow boundaries) at each of the six scale levels (0–5) each lasting 10 seconds were shown. The videos had a visible white printed label describing plate size and SPOT GRADE level. Testing After completing the training, surgeons viewed 108 conse- cutive bleeding video segments (10 s each) each followed immediately by time (10 s) to record the SPOT GRADE level on a score sheet. The testing videos consisted of a second set of different videos previously unseen by the attending sur- geons (i.e., not the training videos) each repeated three times in a random order as described by statistically prepared tables for flow rates in each SPOT GRADE interval and overall video order. Statistics Using these tests, inter- and intra-rater reliability were estimated using a two-way analysis of variance (ANOVA), Figure 3. Apparatus used to create videos of simulated wound bleed- treating surgeons as a random sample from the population ing including wound simulator with interchangeable plates, peristal- of surgeons. In this way, the inter-rater reliability coefficient tic pump system, video camera, and video monitor. was computed as the ratio of the variance component due to Spine www.spinejournal.com 3 Copyright ß 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. CE: A.U.; SPINE154984; Total nos of Pages: 8; SPINE154984 SURGERY The SPOT GRADE  Spotnitz et al TABLE 1. Specific Values for SPOT GRADE Levels Flow Rate (mL/min) Ranges 2 TBS (cm ) SBSS 0 SBSS 1 SBSS 2 SBSS 3 SBSS 4 SBSS 5 1 0 [0;4.8] [4.8; 12.0] [12.0; 25.3] [25.3; 102.0] [102.0; þ1] 10 0 [0;9.1] [9.1; 20.0] [20.0;71.3] [71.3; 147.4] [147.4; þ1] 50 0 [0;13.5] [13.5; 28.0] [28.0;117.3] [117.3; 192.7] [192.7; þ1] SBSS indicates surface bleeding severity scale; TBS target bleeding site. subjects (test videos) over the total variance.26 95% confi- RESULTS dence intervals (CIs) were produced for all inter- and intra- The model input and output blood flow was compared and rater reliability coefficients. All statistical analyses were expressed by calculating a Pearson coefficient (Table 2). For performed using SAS Version 9.3 (Cary, NC). the overall samples, the Pearson correlation coefficient was 0.99280, indicating a strong linear correlation (Figure 5— Determination of Sample Size graph of inflow and outflow) between the measured input The primary sample size for the study was chosen to blood flow rate (flow probe) and calculated blood outflow provide sufficient power for tests of the inter- and intra- (sponge weight change). class correlation coefficients. Landis and Koch27 have For surgeon video testing results (Table 3), inter-rater characterized reliability coefficients as follows: slight reliability was assessed by calculating the interclass correla- (0.0 – 0.20), fair (021 – 0.40), moderate (0.41 – 0.60), sub- tion coefficient (ICCE) ¼ 0.89840 (95% one-sided lower- stantial (0.61 – 0.80), and almost perfect (0.81 – 1.00). limit CI ¼ 0.85771, 1). The null hypothesis was that the Using these divisions as a benchmark, we sought to test intraclass correlation estimating inter-rater reliability was the null hypothesis that the inter-rater reliability coeffi-