The prevalence of inappropriate blood tests in pediatric patients scheduled for elective surgery in Thailand: a retrospective chart review

This study was approved by the Institutional Review Board at Siriraj Hospital (021/2554(EC4)). The medical records of children <15 years old, who were admitted to the surgical ward for elective surgery between May and October 2010, were reviewed retrospectively.

We recorded patient demographic and clinical data including age, sex, their physical status class according to the American Society of Anesthesiologists (ASA), medical history (previous underlying diseases and the present indication for surgery), diagnosis, operation, service unit (categorized by the subspecialty of attending physician and according to nonpediatric or pediatric surgery, which included operations only for pediatric patients), and grade of operation (major surgery was classified as thoracic or abdominal surgery). The preoperative laboratory tests performed for each patient were recorded, including complete blood count (CBC), urinalysis, electrolytes, coagulation studies, and liver function tests (LFT). Results of chest radiographs (CXR) and electrocardiograms (EKG) were also recorded. Each test was determined to be appropriate or inappropriate based on the recommendations listed from the Department of Anesthesiology, Siriraj Hospital, as shown in the Supplementary Table.

We conducted a pilot study to estimate the sample size. According to the pilot study, the prevalence of inappropriate laboratory test requests being made was about 60%. Therefore, based on 95% confidence interval (CI) of prevalence of inappropriate test requests of approximately 60% + 8.5%, a sample of 130 subjects was required. Demographic data were analyzed using descriptive statistics. A Fisher exact test was employed to determine factors associated with inappropriate laboratory tests. P < 0.05 was considered significant.

The reviewed medical records were classified according to surgery service units as pediatric (n = 65) and nonpediatric (n = 65). Patient demographic and clinical data are shown in Table 1.

Most patients were aged between 1 and 8 years (62%), ASA class I (60%), and undergoing elective minor surgery (87%).

The overall prevalence of inappropriate requests for testing was 55.4% (158 out of 285 tests ordered). The prevalence of inappropriate preoperative laboratory test requests is shown in Figure 1. The most frequent inappropriate preoperative investigations were CBC (36.9%), electrolytes (20%), CXR (15.4%), blood urea nitrogen (BUN)/creatinine (Cr) (14.6%), and coagulogram (13.8%). The LFTs were the only laboratory tests that were ordered appropriately.

Figure 1

The prevalence of inappropriate investigations in healthy patients (ASA classifications I and II) was higher than in those with ASA classification III (approximately 50% vs 0%) as shown in Table 2.

Although the prevalence of inappropriate investigations was the highest among children aged between 8 and 15 years (70%, P = 0.13), the prevalence of inappropriate laboratory test requests in the other 2 age groups were also approximately 50%. The prevalence of inappropriate preoperative laboratory investigations increased significantly where patients were to undergo major operations (82%. P = 0.01) and in patients assessed in nonpediatric surgery service units (68%, P = 0.001).

Considering each possibly related factor, CBC was the most frequently ordered inappropriate test, and it correlated significantly with ASA classification I (46%. P = 0.02), minor operation (43%, P < 0.001), and nonpediatric surgery unit (59%, P < 0.001). The prevalence of inappropriate laboratory test orders is shown in Table 3.

The prevalence of testing, except for CXR, was not different among patients in the different age groups. Most frequently, CXRs were ordered inappropriately for patients aged between 8 and 15 years (44%, P < 0.01) and in the nonpediatric surgery service units (31%, P = 0.03). Surprisingly, CXR was not performed in half of the patients undergoing major operations in the pediatric surgery unit (53%, P = 0.003). Coagulograms were ordered inappropriately for the group undergoing major operations (47%, P < 0.001). The cost of unnecessary laboratory investigations in this study is shown in Table 4. It was approximately 19,000 Thai baht for 158 tests of 130 patients.

Preoperative laboratory testing is generally requested to obtain more information about the patient to optimize the preoperative clinical conditions and avoid perioperative complications. However, several investigators demonstrated that in the absence of clinical indications, routine testing rarely detects abnormalities; when abnormal findings are detected, they rarely alter surgical or anesthetic management [7-12].

To evaluate preoperative testing for elective surgery, Turnbull and Buck [8] reviewed charts of 2,570 patients and found that only 2% of the tests were significantly abnormal and only 4 patients benefited from routine laboratory testing. Ideal practice should include selective testing of children after taking a careful history and physical examination [1, 2, 7-13]. The nonselective ordering of preoperative tests leads to higher costs, patient discomfort, and a large number of false-positive results [3].

When comparing the recommendations of our hospital for pediatric preoperative laboratory testing with the review by Kumar and Srivastava [13] and the U.K. National Institute for Health and Care Excellence (NICE) guidelines [14], all three state that no laboratory investigations are needed in healthy children scheduled for elective minor surgery. NICE guidelines recommend preoperative examinations only for children classified as ASA I who are scheduled to undergo either minor or major surgery. For patients with kidney disease, lung disease, or cancer, our recommendations and the report by Kumar et al. [13] suggest to assess CBC, BUN, Cr, electrolyte, and CXR. However, there are some discrepancies between the guidelines, such as a lack of comorbid severity grading.

In this study, we found the most frequently and inappropriately ordered test was CBC, which was ordered most for healthy patients (ASA class I and II) scheduled to undergo minor operations. According to the recommendation of our hospital, it is reasonable to check CBC if the patient’s history is suggestive of underlying anemia, renal failure, cardiac or pulmonary disease, history of bleeding disorder, patients undergoing surgery with expected major blood loss, and if the patient is a former preterm infant with postconceptual age (PCA) < 55 weeks. Roy et al. assessed the cases of 2,000 infants and adolescents <18 years old who underwent minor surgery. The incidence of anemia was less than 2%, and the procedures in 73% of these patients were not cancelled with uneventful anesthesia and surgery [15]. Meneghini et al. also showed that, in healthy children undergoing minor operations, the value of the routine determination of hemoglobin concentration is negligible. The prevalence of severe leukopenia, leukocytosis, and thrombocytopenia were extremely low in healthy patients scheduled to undergo elective surgery [16].

The second most inappropriate laboratory test request was for electrolyte quantification; nevertheless, we recommend that electrolytes be evaluated in patients with renal disease, heart disease, liver disease, and children who have received chemotherapy.

CXRs are recommended for patients with underlying heart or respiratory diseases and patients with expected major blood loss. The present results showed that in half of the patients in the surgery group, CXR was not requested. A study by Sane et al. [17] suggested that the routine preoperative CXR should be performed, whereas Cot [18] and the NICE guidelines [14] do not recommend ordering CXR for major surgery. However, LFT was indicated appropriately for patients with liver disease or those who received chemotherapy.

The prevalence of inappropriate laboratory investigations increased significantly with the severity of operation and in the nonpediatric surgery service unit (unfamiliar with pediatric patients). The coagulogram was the most frequently ordered test before major operations. Suchman and Mushlin [19] and Asaf et al. [20] showed that in healthy patients, without a history of bleeding, diathesis, activated partial thromboplastin time, and prothrombin time (PT), bleeding time did not predict the risk of perioperative bleeding and was not recommend for routine preoperative screening.

Patel et al. [21] sent 1,200 questionnaires to anesthesiologists and surgeons who were members of the Society for Pediatric Anesthesia. They showed that testing for hemoglobin levels was routinely ordered in 27%-48% of patients depending on their age. Compared with our study, the prevalence of inappropriate requests to determine CBC in the pediatric surgery unit was lower, yet it still occurred in 15% of the cases. A possible explanation may be insufficient training of new staff. However, CBC was requested inappropriately more frequently and significantly in the nonpediatric surgery units than in the pediatric surgery unit. Unfamiliarity is a possible reason why physicians did not follow the recommendation of the Department of Anesthesiology, or that ordering routine laboratory tests was easier. Another explanation is that surgeons often worried about procedures being delayed or cancelled if some test results were not obtained [22].

The main limitation of our study was the use of recommendations formulated specifically and solely for Siriraj Hospital. We recognize that a number of guidelines are available and data would differ based on the guidelines chosen. Our data represent findings from a single institution and may not be generalizable to populations in other hospitals and health care systems. The cost analysis was performed based on local hospital billing rates.

In conclusion, the prevalence of inappropriate preoperative testing was high, especially for healthy children scheduled to undergo minor operations. Therefore, it is important to train the surgical staff so that inappropriate preoperative laboratory tests are not requested for healthy pediatric patients who will undergo minor operations.

The recommendation of preoperative laboratory investigation for pediatrics from Siriraj Hospital knowledge management website