a b s t r a c t

Objective: To assess frequency of preoperative hematologic testing in a tertiary care pediatric emergency depart- ment (PED) and how often these values predict clinical outcome or change management decisions.

Methods: Single-center retrospective cohort study in a tertiary-care children’s hospital PED. Patients 0-18 years old, presenting between July 1, 2009-July 1, 2011, ultimately undergoing a surgical procedure within 48 h of pre- sentation were included. Patients were defined as having “preoperative” hematologic assessment if these studies were performed solely because the child was going to the operative suite. Patients who met trauma team activa- tion criteria, underwent neurosurgical procedures, or had laboratory studies performed prior to PED arrival were excluded. The primary outcome was the prevalence of preoperative laboratory assessment.

Results: 528 children were included, of whom 301 (57%) underwent preoperative hematologic laboratory evalu- ations. Of these 301 patients, 115 (38%) had abnormal hematologic parameters, and only 3 (1%) of these patients had their perioperative management changed. One additional child had intraoperative bleeding that required blood products but did not undergo preoperative hematologic assessment. All four children had medical histories that would have identified their risk for perioperative bleeding events.

Conclusion: Preoperative hematologic laboratory assessment occurs frequently in children initially cared for in a

tertiary care pediatric emergency department who subsequently undergo operative interventions. Although age- based abnormal hematologic values are often found, rarely are these abnormalities clinically significant. This study suggests that children cared for in a PED without a history concerning for an increased risk of perioperative bleeding does not require preoperative hematologic assessment.

(C) 2017

Introduction

Children evaluated in our pediatric emergency department (PED) often undergo preoperative hematologic laboratory evaluation to iden- tify patients with increased risk of surgical complications, primarily bleeding. This evaluation often includes an automated blood count (ABC), prothrombin time , activated partial thromboplastin (aPTT), and a blood type with antibody screen (T&S). It has been argued that patients undergoing an elective procedure with a personal or fam- ily history indicative of an abnormality that may lead to surgical compli- cations (e.g. bleeding) should undergo appropriate laboratory testing. However, children lacking such a personal or family history or any

? This work has presented at the following meetings :

?? Society of Hospital Medicine Annual Conference: 3/2015 North Carolina Pediatric Society Annual Meeting: 8/2015 Pediatric Academic Society Annual Meeting: 4/2016

* Corresponding author at: Department of Pediatrics, Duke University Medical Center, Box 102376, USA.

E-mail addresses: [email protected] (C. Woll), [email protected] (P.B. Smith), [email protected] (J.W. Fox).

¹ Present address for Dr. Woll: Department of Pediatric Emergency Medicine, Yale School of Medicine, 100 York St, Suite 1F, New Haven, CT, USA, 06511.

concerning signs on physical examination may be spared this “routine” testing [1-7]. Furthermore, the presence of an abnormality on preoper- ative laboratory tests in and of itself is a poor predictor of operative or postoperative bleeding [8-18].

There are significant costs associated with preoperative laboratory evaluation. At our institution, laboratory charges for this battery of tests exceeds $500. Additional downstream costs may be incurred as many laboratory abnormalities are due to error or transient changes, as opposed to true physiologic abnormalities. These added costs include the monetary cost of additional laboratory investigation and procedural charges, delay in the surgical procedure pending the results of any re- peat tests, and physical/psychological trauma to the child related to the additional blood sampling.

To our knowledge, this topic has not been studied in a pediatric sur- gical population presenting to a PED. Presently, patients who present to our PED needing urgent or emergent surgery frequently undergo preop- erative hematologic testing, dictated primarily by anesthesiology or sur- gical physicians. It is currently unknown how often preoperative labs are drawn in our institution’s PED. Further, it is also unknown how often these preoperative hematologic tests are abnormal in this popula- tion, whether abnormal results could have been predicted by a patient’s

http://dx.doi.org/10.1016/j.ajem.2017.06.055

0735-6757/(C) 2017

history and physical examination, and if abnormal results ultimately alter management decisions (i.e. need for perioperative blood products, delay in operative procedure, hematology consultation prior to the sur- gical procedure). This study aims to provide important information on practices relating to pediatric preoperative hematologic laboratory as- sessment in the PED and whether preoperative laboratory assessment may be averted. This may help us evaluate our current PED practices and provide better and more cost-effective care to future patients.

Methods

Study design

This is a single center retrospective cohort study.

Study setting and population

Pediatric patients (0-18 years old inclusive) who presented to the XXX pediatric emergency department between July 1, 2009, and July 1, 2011, and underwent a surgical procedure within 48 h of presentation were included in this study. Patients who met trauma team activation criteria, underwent neurosurgical procedures, or had labs performed prior to arrival in our PED were excluded from this study. Our PED is a tertiary care, level 1 trauma facility with an average annual volume of 16,000 patients.

Study protocol

Subjects were identified by a data manager within the performance services department using our PED’s operational software (Wellsoft, Somerset, NJ). Institutional Review Board approval was obtained prior to data collection. Medical records were reviewed by one of two inves- tigators (CW or JF) to determine patient demographics, operative proce- dure performed, timing of the procedure relative to emergency care, presence of exclusion criteria, final diagnosis, results of any preopera- tive hematologic studies (ABC, PT, aPTT, T&S), indication for the labora- tory evaluation (solely preoperative or part of diagnostic work-up), past medical and family histories, administration of perioperative blood products, occurrence of perioperative hematology subspecialty consul- tation, and occurrence of intraoperative/postoperative Bleeding complications.

A patient who had the above blood tests performed solely for the purpose of surgical bleeding risk assessment was defined as having un- dergone preoperative hematologic laboratory testing (e.g. healthy child without active bleeding had an ABC, PT, aPTT and T&S performed prior to going to the operating room [OR] for percutaneous immobilization of a displaced humeral supracondylar fracture). It is commonplace at our institution for all of these labs to be ordered in this bundle.

Outcome measures

The primary outcome was the frequency in which preoperative he- matologic laboratory testing was performed. Secondary outcomes in- cluded the frequency of abnormal laboratory values, frequency of alterations in a patient’s clinical course as a result of abnormal laborato- ry values (i.e. delay in the surgical procedure, administration of blood products perioperatively, pediatric hematology subspecialty consulta- tion), and the frequency of perioperative bleeding complications. Labo- ratory values were deemed abnormal per previously published guidelines [19,20].

Data analysis

Descriptive statistics were performed on this dataset: percentage of patients who had preoperative labs are performed, percentage of those patients with laboratory abnormalities, percentage of patients who

required perioperative blood products, percentage of patients who re- quired pediatric hematology subspecialty consultation in the periopera- tive period, and percentage of patients with perioperative bleeding complications. Stata 13 (College Station, TX) was used for all analyses.

Results

A total of 947 patient charts were initially evaluated. We excluded 231 patients admitted to a surgical service from the PED but not taken to the OR during that admission, 108 patients who met trauma team ac- tivation criteria, 55 neurosurgical procedures, 10 children taken to the OR more than 48 h after care in the PED, and 15 charts with missing data (Fig. 1). We included children up to 48 h of PED presentation as we felt procedures performed in this time frame were most likely relat- ed to the reason for PED presentation. Ultimately, 528 children were in- cluded in our study population. The average age of the study population was 8.1 years and 35% were female (Table 1). Of these, 301 (57%) underwent preoperative hematologic laboratory evaluation. Of these 301 children, 115 (38%) had abnormal hematologic values. The break- down of the abnormal laboratory results was as follows: 63 (55%) had isolated prolonged PT or aPTT, 29 (25%) had isolated anemia, 2 (1.7%) had isolated thrombocytopenia, 18 (16%) had prolonged PT or aPTT with anemia, and 2 (1.7%) had prolonged PT or aPTT with thrombocyto- penia, and 1 (0.9%) had anemia with thrombocytopenia. Table 2 shows the 5 most common procedures performed in patients with and without preoperative laboratory assessment.

Of the 528 children in this study population, only 4 (0.8%) experi- enced an a priori defined change in their perioperative course; each re- quiring perioperative blood products without a delay in the planned surgical procedure or pediatric hematology Subspecialty consultation. Three of these children had undergone preoperative hematologic as- sessment while 1 did not. All 4 children had medical histories indicating increased risk for perioperative bleeding complications (Table 3).

There were only 19 (3.6%) children in our cohort with past medical histories that would raise concern for perioperative bleeding complica- tions. Examples include idiopathic thrombocytopenic purpura, complex congenital heart disease requiring Warfarin therapy, inflammatory bowel disease, and short shut syndrome. As noted above, all 4 children who required blood products would have been identified by their past medical histories and thus warranted preoperative hematologic labora- tory assessment. Importantly, none of the remaining 509 (96.4%) chil- dren without concerning medical histories required perioperative blood products, underwent pediatric hematology subspecialty evalua- tion, or had an unexpected delay in their surgical procedure as a result of abnormal laboratory values. Finally, a family history of a bleeding dis- order was not predictive of a child’s need for blood products in our study population (data not shown).

Discussion

According to our single-center cohort, pediatric patients taken to the operating room after initial care in a pediatric emergency department without medical histories concerning for increased perioperative bleed- ing are at Very low risk for perioperative bleeding complications. Preop- erative hematologic laboratory assessment is common in our emergency department with more than half patients undergoing this evaluation. This high prevalence is surprising given the substantial pub- lished literature indicating that this work-up is not necessary in children without concerning hematologic histories [1-7] and the overall small percentage of children in our study with high risk features. Importantly, none of the children (96% of the study population) without worrisome past medical histories as they relate to bleeding risk required unexpect- ed perioperative interventions due to bleeding (defined here as requir- ing perioperative blood products, perioperative pediatric hematology subspecialty consultation, or delay in the surgical procedure due to bleeding).

Fig. 1. Patient flow chart Abbreviations: OR = operating room, ED = emergency department.

Despite the increasing focus on cost conscious and patient-centered care by families, medical providers, and health care payers, there is a paucity of information regarding the utility of preoperative laboratory evaluation in the PED. This study showed that despite frequent preoper- ative hematologic assessments, rarely did this testing affect a patient’s operative course. Furthermore, when preoperative laboratory studies were obtained, abnormal parameters were frequently found yet the vast majority of these did not result in any clinically significant adverse events. In fact, all children in our study who required perioperative blood products had medical histories that would have alerted the med- ical team to the need to obtain preoperative hematologic assessment. Our study is congruent with published literature in other pediatric clin- ical realms which state that abnormal preoperative hematologic studies are poor predictors of operative bleeding complications [8-18] and thus a patient’s personal history (i.e. documented Bleeding disorder, on anticoagulation therapy, active bleeding) should drive the decision to obtain preoperative hematologic testing [1-7].

Limitations

We are limited by the single center and retrospective nature of the study. By gathering information from a single center, we limit the

generalizability of the conclusion. However, including 528 children in the study provides a relatively large sample size which augments the limited generalizability of the single-center approach. We conducted this study in a retrospective manner in order to collect epidemiological information regarding our current practices and how those practices

Table 1

Study population demographics.

	Overall		Labs NOT obtained		Labs obtained
	n = 528		n = 227		n = 301
Age (mean in years)	8.1		6.7		9.2
Female (%) Race/ethnicity (%) African-American	187 (35) 142 (27)		85 (37) 68 (30)		102 (34) 74 (25)
American Indian	2 (0.4)		1 (0.4)		1 (0.3)
Asian	9 (2)		4 (2)		5 (2)
Caucasian	221 (42)		94 (41)		127 (42)
Hispanic/Latino	75 (14)		26 (11)		49 (16)
Multiracial	8 (2)		2 (1)		6 (2)
Not recorded	71 (13)		32 (14)		39 (13)
Hematological significant	19 (3.6)		4 (1.8)		15 (5)

history (%)

Table 2 Top 5 procedures performed for patients who did and did not have preoperative hemato- logic laboratory assessment.

Conclusions

Preoperative hematologic laboratory assessment occurs frequently

Pre-operative labs obtained (n = 301)

Pre-operative labs NOT obtained (n = 227)

in children initially cared for in a tertiary care pediatric emergency de- partment who subsequently undergo operative interventions within

Procedure n (%) Procedure n (%) Internal fixation of fracture 101 (34) Appendectomy 61 (27)

48 h of emergency care. Although age-based abnormal hematologic values are often found, rarely are these abnormalities clinically signifi-

Appendectomy 65 (22) incision and drainage of

Cutaneous abscess

42 (19)

cant. This study suggests that children cared for in a PED without a his-

tory concerning for an increased risk of perioperative bleeding do not

Wound/laceration repair 31 (10) Foreign body removal 36 (16)

Incision and drainage of cutaneous abscess

23 (8) Wound/laceration repair 23 (10)

require preoperative hematologic assessment.

Exploratory laparotomy 16 (5) fracture reduction and

immobilization

?Procedures listed in descending order of frequency for both categories.

19 (8)

Acknowledgements

Thank you to Katie Westreich, MD, who obtained initial IRB approval and developed the data collection template. This research did not re- ceive any specific grant from funding agencies in the public, commer- cial, or not-for-profit sectors.

affect our management. This methodology does not prove causality but only demonstrates association which is sufficient for epidemiologically- focused studies. Future studies should focus on increasing the sample size, expanding the study population to multiple centers, conducting re- search in a prospective fashion, and studying the role of platelet func- tion in the child requiring urgent/emergent surgery.

We view our study as an opportunity to reduce waste in our PED. The vast majority of children who present to our PED carry low risk for perioperative bleeding complications. Currently, more than half of the children who are initially cared for in our PED then undergo surgical intervention within 48 h have preoperative hematologic testing per- formed. If we can limit this testing to children only at higher risk for perioperative bleeding, we stand to reduce testing by up to 93%. As such, we plan to work with our surgical and anesthesiology colleagues to develop and implement an algorithm in our PED to reduce unneces- sary preoperative hematologic laboratory evaluations. We hope to study the effect this algorithm has on Medical costs and patient outcomes.

References

1. Abbasi S, Khan FA. Effect of pre-operative coagulation testing on intra-operative Transfusion requirements in surgical patients. J Coll Physicians Surg Pak 2005;15: 319-22.
2. Almesbah F, Mandiwanza T, Kaliaperumal C, Caird J, Crimmins D. Routine preopera- tive blood testing in pediatric neurosurgery. J Neurosurg Pediatr 2013;12:615-21.
3. Asaf T, Reuveni H, Yermiahu T, et al. The need for routine pre-operative coagulation Screening tests (prothrombin time PT/partial thromboplastin time PTT) for healthy children undergoing elective tonsillectomy and/or adenoidectomy. Int J Pediatr Otorhinolaryngol 2001;61:217-22.
4. Derkay CS. A cost-effective approach for preoperative hemostatic assessment in chil- dren undergoing adenotonsillectomy. Arch Otolaryngol Head Neck Surg 2000;126: 688.
5. Eroglu E, Sozmen BO, Kayiran SM, Balci S, Gurakan B. Evaluation of coagulation tests before newborn circumcision: is it necessary? Blood Coagul Fibrinolysis 2016;27: 160-2.
6. Licameli GR, Jones DT, Santosuosso J, Lapp C, Brugnara C, Kenna MA. Use of a preop- erative bleeding questionnaire in pediatric patients who undergo adenotonsillectomy. Otolaryngol Head Neck Surg 2008;139:546-50.
7. Manning SC, Beste D, McBride T, Goldberg A. An assessment of preoperative coagu- lation screening for tonsillectomy and adenoidectomy. Int J Pediatr Otorhinolaryngol 1987;13:237-44.

   Table 3

   Patients requiring perioperative blood products.

   Patient description Preoperative labs

   obtained

   Laboratory abnormality

   11-year-old male with a history of hypoplastic left heart syndrome status post Fontan procedure and aortic coarctation status post balloon dilatation was anticoagulated with warfarin required laparoscopic appendectomy for acute appendicitis. His preoperative hematologic assessment was notable for a prolonged PT/supratherapeutic INR (33.5 s / 2.8). He was given fresh frozen plasma prior to his operation. He did not experience any perioperative complications.

8. 17-year-old male with Crohn’s disease complicated by chronic anemia required operative management of a Small bowel obstruction. His preoperative hematologic assessment was notable for anemia with a hematocrit of 9.7 mg/dl (baseline 10-11 mg/dl). On his third postoperative day, his hematocrit had declined to 6.3 mg/dl. The patient experienced symptoms due to this significant anemia, so he was transfused Packed red blood cells.
9. 4-year-old female with a history of autoimmune encephalitis undergoing chronic plasmapheresis required operative replacement of a broken Broviac catheter (removal of broken catheter and placement of a new catheter). Preoperative hematologic laboratory assessment was not performed. While in the OR, there was prolonged bleeding at the surgical site for the new catheter despite multiple attempts at holding direct pressure for 5-min time intervals. An automated blood count, PT/INR, aPTT, and Fibrinogen levels were measured intraoperatively. She had a mild anemia (10.0 g/dl - patient’s baseline), slightly prolonged PT (13.6 s)/INR (1.2), prolonged aPTT (45.6 s), and low fibrinogen (81.5 mg/dl). Bleeding was controlled after the patient was given DDAVP and cryoprecipitate with fibrin glue applied to the wound. She experienced approximately 60 ml blood loss. The remainder of her perioperative course was uneventful.
10. 12-year-old previously healthy male presented to the PED with 2-3 days of frank hematochezia. His hemoglobin in the PED was

    11.3 g/dl. He was admitted overnight with a plan for gastrointestinal endoscopy the following morning. His bleeding continued overnight. His endoscopy was delayed so that he could receive pRBCs for an acute decrease in his hemoglobin to 7.6 g/dl. Colonoscopy revealed a large amount of blood in his entire colon with the source seemingly proximal to the terminal ileum. His Upper endoscopy was normal. Meckel’s scan was negative but a tagged RBC scan localized active bleeding in his left lower abd. Angiography confirmed bleeding in the rectosigmoid region. He was taken to the OR the following day where a mucosal arteriovenous malformation was found in his rectum. This was over sewn with sutures with hemostatic control obtained. He had required several pRBC transfusion as well as FFP during his hospitalization. However, bleeding stopped after the AVM was identified and treated in the OR.

    Yes Elevated PT/INR

    Yes Anemia

    No Elevated

    PT/INR/aPTT,

    anemia,

    low fibrinogen

    Yes Anemia

    * PT = prothrombin time, INR = international normalized ratio, aPTT = activated partial thromboplastin time, OR = operating room, DDAVP = deamino-delta-D-arginine vasopressin, PED = pediatric emergency department.

    Close HL, Kryzer TC, Nowlin JH, Alving BM. Hemostatic assessment of patients before tonsillectomy: a prospective study. Otolaryngol Head Neck Surg 1994; 111:733-8.

11. Cooper JD, Smith KJ, Ritchey AK. A cost-effectiveness analysis of coagulation testing prior to tonsillectomy and adenoidectomy in children. Pediatr Blood Cancer 2010; 55:1153-9.
12. Eisert S, Hovermann M, Bier H, Gobel U. Preoperative screening for coagulation dis- orders in children undergoing adenoidectomy (AT) and tonsillectomy (TE): does it prevent bleeding complications? Klin Padiatr 2006;218:334-9.
13. Eley KA, Goodacre TE. Are routine pre-operative blood tests required in children un- dergoing primary cleft lip and/or palate repair? J Plast Reconstr Aesthet Surg 2010; 63:926-9.
14. Gabriel P, Mazoit X, Ecoffey C. Relationship between clinical history, coagulation tests, and perioperative bleeding during tonsillectomies in pediatrics. J Clin Anesth 2000;12:288-91.
15. Houry S, Georgeac C, Hay JM, Fingerhut A, Boudet MJ. A prospective multicenter evaluation of preoperative hemostatic screening tests. The French associations for surgical research. Am J Surg 1995;170:19-23.
16. Koscielny J, Ziemer S, Radtke H, et al. A practical concept for preoperative identifica- tion of patients with impaired primary hemostasis. Clin Appl Thromb Hemost 2004; 10:195-204.
17. Mallick MS. Is routine pre-operative blood testing in children necessary? Saudi Med

    J 2006;27:1831-4.

    Munro J, Booth A, Nicholl J. Routine preoperative testing: a systematic review of the evidence. Health Technol Assess 1997(1):1-62 (i-iv).

18. Narr BJ, Hansen TR, Warner MA. Preoperative laboratory screening in healthy Mayo patients: cost-effective elimination of tests and unchanged outcomes. Mayo Clin Proc 1991;66:155-9.
19. Segal JB, Dzik WH. Paucity of studies to support that abnormal coagulation test re- sults predict bleeding in the setting of invasive procedures: an evidence-based re- view. Transfusion 2005;45:12.
20. Andrew M, Vegh P, Johnston M, Bowker J, Ofosu F, Mitchell L. Maturation of the he- mostatic system during childhood. Blood 1992;80:1998-2005.
21. Appel IM, Grimminck B, Geerts J, Stigter R, Cnossen MH, Beishuizen A. Age depen- dency of coagulation parameters during childhood and puberty. J Thromb Haemost 2012;10:2254-63.