Impact of using a local protocol in preoperative testing: blind randomized clinical trial

Objective: to evaluate the impact of the use of a local protocol of preoperative test requests in reducing the number of exams requested and in the occurrence of changes in surgical anesthetic management and perioperative complications. Methods: we conducted a randomized, blinded clinical trial at the Gaffrée and Guinle University Hospital with 405 patients candidates for elective surgery randomly divided into two groups, according to the practice of requesting preoperative exams: a group with non-selectively requested exams and a protocol group with exams requested according to the study protocol. Studied exams: complete blood count, coagulogram, glycemia, electrolytes, urea and creatinine, ECG and chest X-ray. Primary outcomes: changes in surgical anesthetic management caused by abnormal exams, reduction of the number of exams requested after the use of the protocol and perioperative complications. Results: there was a significant difference (p<0.001) in the number of exams with altered results between the two groups (14.9% vs. 29.1%) and a reduction of 57.3% in the number of exams requested between the two groups (p<0.001), which was more pronounced in patients of lower age groups, ASA I, without associated diseases and submitted to smaller procedures. There was no significant difference in the frequency of conduct changes motivated by the results of exams or complications between the two groups. In the multivariate analysis, complete blood count and coagulogram were the only exams capable of modifying the anesthetic-surgical management. Conclusion: the proposed protocol was effective in eliminating a significant number of complementary exams without clinical indication, without an increase in perioperative morbidity and mortality.


INTRODUCTION
T he preoperative evaluation (POE) seeks to promote safety in surgery and anesthesia, to ensure a better quality of care, as well as the rational use of resources in the perioperative period.Thus, history and physical examination should be considered the main components of the POE, with the complementary exams remaining under specific clinical conditions [1][2][3] .
In general, patients who are candidates for elective operations have requested preoperative complementary exams (POCE) routinely and indifferently to the clinical findings of the POE.This is based on several factors, such as: ability to identify diseases not diagnosed by anamnesis and physical examination, safety assurance to the professionals involved in the process to make decisions regarding the resolution of intercurrences, as well as safeguarding possible legal responsibilities 4 .However, the medical literature has indicated that abnormalities found in POCEs are not usually clinically important, are generally ignored, do not contribute to changes in the anesthetic-surgical management and are not related to perioperative complications.In addition, there is a risk that non-clinically based tests, especially on the occurrence of false positives, may lead to further invasive investigations, leading to postponement of operations as well as inadequate treatment 3,5 .
Regarding selected exams, more controlled clinical research is needed 6 .The rationalization of the request for complementary exams in the POE still requires studies, and to this end, emerged evidence-based guidelines 1,2,[7][8][9] and protocols constructed in view of presence of associated diseases and the procedures to be performed 1,2 .Some authors show that the implementation of protocols increases the effectiveness of the requests for exams without affecting patient's safety and the morbidity of the surgical-anesthetic procedure 10,11 .
1 -Gaffree and Guinle University Hospital, Federal University of the State of Rio de Janeiro, Service of Anesthesiology, Rio de Janeiro, Rio de Janeiro State, Brazil. 2 -School of Medicine and Surgery, Federal University of the State of Rio de Janeiro, Service of General and Specialized Surgery, Rio de Janeiro, Rio de Janeiro State, Brazil.

Original Article A B S T R A C T
Objective: to evaluate the impact of the use of a local protocol of preoperative test requests in reducing the number of exams requested and in the occurrence of changes in surgical anesthetic management and perioperative complications.Methods: we conducted a randomized, blinded clinical trial at the Gaffrée and Guinle University Hospital with 405 patients candidates for elective surgery randomly divided into two groups, according to the practice of requesting preoperative exams: a group with non-selectively requested exams and a protocol group with exams requested according to the study protocol.Studied exams: complete blood count, coagulogram, glycemia, electrolytes, urea and creatinine, ECG and chest X-ray.Primary outcomes: changes in surgical anesthetic management caused by abnormal exams, reduction of the number of exams requested after the use of the protocol and perioperative complications.Results: there was a significant difference (p<0.001) in the number of exams with altered results between the two groups (14.9% vs. 29.1%)and a reduction of 57.3% in the number of exams requested between the two groups (p<0.001), which was more pronounced in patients of lower age groups, ASA I, without associated diseases and submitted to smaller procedures.There was no significant difference in the frequency of conduct changes motivated by the results of exams or complications between the two groups.In the multivariate analysis, complete blood count and coagulogram were the only exams capable of modifying the anesthetic-surgical management.Conclusion: the proposed protocol was effective in eliminating a significant number of complementary exams without clinical indication, without an increase in perioperative morbidity and mortality.The objective of this study was to evaluate the use of a local protocol of preoperative examination requests in the POE, including its impact on the number of requested exams, the occurrence of changes in anesthetic-surgical management and the frequency of perioperative complications.The POE consultations were performed at least 15 days before the operation and followed the guidelines of the American Society of Anesthesiologists (ASA) 1 .

METHODS
The study participants were randomly divided by simple randomization into: Routine Group (RG), in which the routine preoperative complementary exams were requested before POE consultation, and Protocol Group (PG), whose request for exams was Based on the POE consultation following the Preoperative Exam Request Protocol (Figure 1), developed by the researcher according to the guidelines of the ASA Task Force 1 .
The POCE included in this study were complete blood Other variables studied were the total number of exams requested, the number of exams with abnormal results, the number of exams additional to the protocol, and the difference between the total and individualized number of exams requested between the two groups.
The sample calculation was performed based on the historical average of the last 24 months of the number of patients seen at the POE/HUGG Clinic and considered  RG and PG, respectively.The frequency of altered results was higher in the PG for CBC (p=0.009),serum urea and creatinine concentration (p=0.005),PT/aPTT (p=0.008),blood glucose (p<0.001), and chest X-ray (p<0.001).
Serum electrolyte concentration and ECG did not reveal statistical significance between the groups (Table 2).The PG underwent less POCE than the RG (p<0.001) for all types of exams except for the category supplementary exams (p=0.158)(Figure 2).The POCE mean of the PG was  3).Considering the change in conduct and the POCE, we observed statistically significant associations, and the chance of conduct change was 8.48 times higher for the altered blood count when compared to the normal CBC.
For the evaluation of the PT/aPTT, this estimate was even higher (OR=30.28,95% CI= 5.17-177.55).However, one must exercise caution with this finding because of the size of the confidence interval (Table 4).
The frequency of complications was 58 cases in the RG (28.43%) and 54 cases in the PG (26.86%), with p=0.658 (Table 3).There was an increase in the risk estimates in the association between the characteristic "associated diseases" and the occurrence of the complication "hypotension", with a cumulative effect according to the increase in the number of associated diseases, with a statistically significant result (OR=3.51,95% CI= 1.41-8.73)(Table 4).We observed a positive association between the ASA variable and the other cardiovascular complications.Nonetheless, the values found were not statistically significant.When the association between this group of complications and the classification of MET was evaluated, we observed that individuals classified as MET's ≤4 presented a three times greater chance of complications when compared with the group classified with MET's ≥4 (Table 4).

DISCUSSION
Several studies seek to analyze the effectiveness of preoperative examinations in modifying surgical anesthetic management and its impact on the frequency of perioperative complications.One of the pioneer studies was a retrospective study from Kaplan et al. 12 , which concluded that only 4.3% of the routine exams had abnormal results and of those, 0.17% had some clinical significance.Similarly, Soares et al. 5 , in a descriptive study, found 2.25% of changes in the results of such exams and only 0.38% were cause for behavior change.
Differently from these findings, our study revealed that in the RG group 14.9% of the exams were altered and there was a change in surgical anesthetic management in 2.5% of operations.These data are similar to those found by Guerra et al. 13 , who conducted a retrospective study and observed 19.8% of abnormalities in routine preoperative exams, which led to a change of conduct in 1.8% of cases.Benarroch-Gampel et al. 14 , in a retrospective cohort using data from the American National Surgical Quality Improvement Program of patients with a profile similar to that of our sample, found that 61.6% had at least one abnormal test result.Roizen 15 observed that, assuming that the tests are independent of each other, the more tests are requested, the greater the likelihood of finding abnormal results if we take into account issues such as the specificity and sensitivity of such tests and that routine tests are not suitable instruments to identify unknown diseases, are not cost effective, have no impact on surgical anesthetic management, and potentially add risks to the patient and medical-legal problems.
Regardign selective exams, they were altered in 29.1% of our study.Charpak et al. 16 , in an analysis of surgical patients who underwent a chest X-ray with a clinical indication identified that 52% had abnormalities in the examination, but in only 23% of the cases, these abnormalities were not expected by anamnesis and physical examination.Also in this study, the observed changes modified the anesthetic-surgical management in only 5% of the cases.Chung et al. 17 , in a randomized clinical trial of patients undergoing outpatient surgical procedures, found 11.5% of selective exams with abnormal results and only one change in perioperative care was detected.In these studies, it can be concluded that abnormalities found in selective preoperative exams can often be anticipated by anamnesis and physical examination and/or reflect physiological changes in aging, and have no impact on surgical planning and anesthetic management, such as observed in our study.
These findings have implications not only on the direct and indirect financial costs of this practice, but also on the quality of the preoperative evaluation.Thus, we share the opinion of Roizen 18 , which says that the real question is not to do or not to do exams, but rather to establish the clinical characteristics of the surgical patient through a careful anamnesis and physical examination and not to replace them by the complementary exams requests.
In the last 20 years several medical societies and health technology institutes 1,2,7-9 have developed protocols with the purpose of reducing the routine of requesting non-selective preoperative exams, a practice that found no evidence support.One of the first studies was that of Fischer et al. 19 , who compared patients in preoperative elective surgery distributed in two groups: in one group the tests were routinely ordered and in the other they were based on the presence of specific clinical conditions.These authors observed a reduction of 55.14% in the number of exams requested without there being an increase in the procudures' cancellations.
Similarly, the present study found a reduction of 57.3% in the requests for preoperative exams between the study groups.This reduction ranged from 28.8% for ECG to 89.6% for PT/aPTT.When analyzing only the exams done by the PG group, we found that 22.4% of based guidelines.In this study, there was no reduction in the average number of exams found, complications were higher in the group of exams selected, but these were not related to the exams' results or to their absence.In the present study, when submitting a group of patients to the proposed protocol, we observed a reduction in the mean number of exams requested, similar to the findings of Mancuso 9 and Barazzoni 10 , that is, a more pronounced reduction of exams requests for those younger patients, without associated diseases, classified as ASA I and undergoing small operations (procedures with local anesthesia and sedation).This significant reduction in the number of requested exams was not accompanied by increase in the incidence of perioperative complications or in the request for exams complementary to the protocol.

R E S U M O
This is a blinded, randomized clinical trial conducted at the Gaffrée and Guinle University Hospital (HUGG) of the Federal University of the State of Rio de Janeiro (UNIRIO), between March 2014 and July 2015, and approved by the Ethics in Research Committee under the Number 27505514400005258.All patients signed an Informed Consent Form.The population comprised patients who were candidates for surgical procedures in General Surgery, Digestive Surgery and Coloproctology (here grouped as General Surgery), Urology, Gynecology, Thoracic Surgery, Vascular Surgery, Otorhinolaryngology, Orthopedics, Neurosurgery, Plastic Surgery and Ophthalmology.The inclusion criteria were age greater than or equal to 18 years, elective operation, and preoperative evaluation conducted at the HUGG Preoperative Evaluation Clinic.Exclusion criteria were age less than 18 years, emergency/ urgency operations, ASA IV or V and those whose preoperative evaluation was done in another hospital.
count (CBC), prothrombin (PT and INR) and activated partial thromboplastin time (aPTT), serum electrolytes concentration (sodium, potassium and chlorine), glucose, urea and creatinine, resting electrocardiogram (ECG) and chest radiography.All data on history, physical examination and test results were recorded on individual preoperative assessment sheets in addition to data from complementary exams that were not part of the protocol, but were deemed necessary and requested by the POE conductor (supplementary examinations).On the day of the surgical procedure, after discharge from the post anesthetic recovery (PAR) room, all patients underwent an evaluation by anesthesiologists responsible for the anesthetic procedure with observation and recording of the following parameters: alteration in surgical anesthetic management (cancellation of the procedure, change in anesthetic and/or surgical technique or change in postoperative care) due to absence or abnormal results of preoperative exams (outcome 1) and/or complications, during the anesthetic-surgical procedure or in the period between the patient leaving the operating room until discharge from the PAR room (outcome 2).Complications considered were hypotension (systolic blood pressure ≤80mmHg), cardiac arrhythmia in a patient with no previous history or worsening of preexisting disorder requiring treatment, hypertension (systolic blood pressure ≥200mmHg or diastolic BP≥110mmHg) and cardiorespiratory arrest.These last three were grouped, for statistical purposes, as other cardiovascular complications.Respiratory complications were thus grouped: hypoxemia (SATO 2 ≤90% or PaO 2 ≤100mmHg), laryngospasm, bronchospasm, thoracic stiffness, residual curarization and difficulty of orotracheal intubation (OTI).Shock, regardless of the cause, was also included, and a group of general complications (nausea and vomiting, inadequate pain control, prolonged awakening, agitation on awakening, hypoglycemia, and total or partial block failure).

2 .
98±2.04.Fifty additional exams were requested, 29 in the routine group and 21 in the protocol group, with no statistical difference between the groups.There were conduct changes caused by absence or altered outcome of the exams in seven surgical procedures (1.8% of operations), with five cases (2.5%) in the RG and two cases (1.0%) in the PG (p=0.231)(Table

Figure 2 .
Figure 2. Reduction of the number of exams with the Protocol.

Table 1 .
Demographic and clinical data.

Table 2 .
Distribution of results by analysis group.

Table 3 .
Conduct changes and complications.

Table 4 .
Influence of exams on surgical-anesthetic conduct and of the sample characteristics on operative complications.
PT/aPTTÇ prothrombin time and activated partial thromboplastin time; ASA: physical status according to the American Society of Anaesthesiologists; Met: metabolic equivalents (activity index of Duke).