Anesth Analg 2002;95:1563-1568
© 2002 International Anesthesia Research Society
CARDIOVASCULAR ANESTHESIA
The Effect of Alterations in a Preoperative Assessment Clinic on Reducing the Number and Improving the Yield of Cardiology Consultations
Lawrence C. Tsen, MD*,
Scott Segal, MD*,
Margaret Pothier, CRNA*,
L. Howard Hartley, MD
, and
Angela M. Bader, MD*
*Departments of Anesthesiology, Perioperative and Pain Medicine and Internal Medicine, Harvard Medical School, Brigham and Women’s Hospital, Boston, Massachusetts
Address correspondence and reprint requests to Angela M. Bader, MD, Harvard Medical School, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Address e-mail to ambader{at}bics.bwh.harvard.edu
 |
Abstract
|
|---|
Although preoperative assessment testing clinics (PATCs) can produce efficiency in the evaluation of surgical candidates, their effect on the use of consultants has not been studied. We hypothesized that changes in PATC procedures, education, and staffing could affect the use and yield of cardiology consultations. All PATC anesthesiologist-requested cardiology consultations for patients undergoing elective noncardiac surgery from 1993 to 1999 were reviewed. This period corresponded to 3 yr before and after a change in the PATC leadership, which resulted in more stringent consultation algorithms, a cardiac assessment and electrocardiogram interpretation educational program, and altered staffing of anesthesiologists and ancillary personnel. A single senior cardiologist completed all consultations. Data including age, sex, reason for consultation, resultant testing, consultant conclusions, cancellations, and surgical procedure and outcomes were collected. In the PRE and POST groups, respectively, 917 and 279 consultations (1.46% versus 0.49% [P = 0.0001] of noncardiovascular surgeries) were ordered despite an increase in the surgical case-mix acuity. In the POST group, significantly fewer consultations were ordered and significantly more required further testing to assess cardiac status. We conclude that changes in PATC consultation algorithms, education, and staffing can significantly decrease the use and yield of preoperative cardiology consultations.
IMPLICATIONS: Alterations in preoperative assessment testing clinic consultation algorithms, education, and staffing can significantly reduce the use of preoperative cardiology consultations while improving their overall yield.
|
Introduction
|
|---|
Providing high-quality, cost-effective preoperative evaluation of surgical patients is a key issue in perioperative patient management (1). In the patient with known or suspected cardiac disease undergoing noncardiac surgery, the assessment process remains controversial, with some investigators suggesting that routine presurgical anesthesia assessments are neither sufficiently sensitive nor specific enough for risk assessment to occur (2,3). Therefore, although risk factors for perioperative cardiac morbidity have been identified (4,5), evaluation is often made in consultation with a cardiologist. Confusion, however, by the anesthesiologist and the cardiologist as to the overall purpose and utility of the consultation is commonplace (6), with few anesthesiologists specifying the reason for the consultation or indicating questions to be answered (7,8). In addition, because "medical-legal considerations" (5) or "clearance for surgery" (6) are often cited by anesthesiologists as the purpose for the consultation, a number are performed on patients with known, stable medical disease or result in few changes in preoperative therapies or interventions. It is not surprising that few anesthesiologists ultimately feel obliged to follow the consultant’s recommendations (5).
In our institution, preoperative consultations are ordered by attending anesthesiologists staffing the preoperative assessment testing clinic (PATC) and performed by a single senior attending cardiologist. Preliminary investigation suggested that a seemingly limited number of completed consultations resulted in a diagnosis of new or unstable cardiac disease or required additional cardiologist-ordered testing before surgery. Therefore, we instituted prominent procedural, educational, and staffing alterations within our high-volume tertiary-care PATC and examined the effect on patterns of cardiology consultation, resulting interventions, and patient outcomes before and after the change. We hypothesized that improvements in the appropriateness and yield of consultations could be achieved as a result of those changes.
|
Methods
|
|---|
After approval by the hospital’s Committee for the Protection of Human Subjects, we performed a retrospective review of all patients receiving anesthesiologist-generated cardiology consultations through the hospital’s PATC during the years 1993 through 1999. This time period corresponded to 3 yr before (1993–1996 [PRE group]) and 3 yr (1997–1999 [POST group]) after a sentinel change in the PATC leadership, which resulted in the implementation of more stringent consultation algorithms, a cardiac assessment and electrocardiogram (ECG) educational program, and altered staffing of attending anesthesiologists and ancillary personnel. The method of cardiology consultations in the preoperative clinic was as follows. After surgical and nursing staff assessments and the completion of indicated basic laboratory and ECG testing, patients were interviewed and examined by an attending anesthesia provider who, at his or her discretion, requested a cardiology consultation. All consultations were fulfilled by a single, senior, board-certified attending cardiologist. Patients scheduled for cardiovascular surgery for whom cardiology consultations were routinely requested by the surgeon before their PATC visit were excluded from this study.
A standardized protocol was used to evaluate and record the data (Table 1). Records reviewed included the cardiology consultation and subsequent testing results, the anesthesia preoperative evaluation and operative record, the recovery room record, and notes from the postoperative course. Data on age, sex, reason for consultation, cancellations, surgical procedures, and overall outcomes were also collected. To indicate the severity of preoperative patient comorbidity and surgical case complexity, the Case Mix Index (CMI) was recorded as an aggregate for all patients undergoing surgery during the PRE and POST time periods. The CMI is a standard, diagnostic related group-based weighting system that evaluates the existence of comorbid disease on the basis of International Classification of Diseases (9th revision) diagnosis codes at the time of discharge. It is specific to the treatment for a particular hospitalization. Descriptive statistical analyses were performed; 
2 testing was used where appropriate, with P < 0.05 considered significant.
|
Results
|
|---|
A total of 1196 consultations were performed and evaluated. A significant decrease in consultations, from 917 to 279, was noted between the PRE and POST groups; this accounted for 1.46% and 0.49% (P = 0.0001) of the total noncardiovascular surgeries performed during their respective time periods. The breakdown by type of surgery is listed in Table 2. Patients scheduled for thoracic and orthopedic procedures accounted for the largest percentage of requested consultations in both groups. In addition, the three most common reasons triggering a consultation in both groups were, in descending order, ECG abnormalities, perceived changes in known cardiac disease, and symptoms suggestive of cardiac disease (Table 3). Significantly fewer consultations were requested in the POST group for ECG abnormalities alone (22.7% versus 49.6%; P < 0.0001).
Demographically, men accounted for a larger percentage of consultations in the POST group (137 [53%] versus 395 [43%]; [P = 0.006]), and the overall mean age ± SD for both sexes was older in the POST group (67.5 ± 11.6 yr versus 63.8 ± 13.3 yr; P = 0.0001). The mean CMI increased in the POST group (analysis of variance; P < 0.0001) for all surgical services (Table 4). This finding is reflected in the larger number of patients in this group being scheduled for hospital admission (n = 210 [81%] versus n = 619 [72%]; P = 0.002) instead of day surgery. When admitted, however, no differences were noted between the groups in length of hospital stay (6.1 ± 7.4 versus 6.3 ± 11.0 days; P = 0.69).
Although no significant differences were noted between groups in the overall impression of the consultant, with the majority indicating minimal or stable known cardiac disease, significantly more interventions were performed in the POST group to achieve this impression (Table 5). Patients in the POST group underwent significantly more exercise stress testing (12.2% versus 27.2%), echocardiograms (7.9% versus 14.8%), or both (1.3% versus 6.2%). Two patients in the PRE group underwent cardiac catheterization to further evaluate their risk; no patients in the POST group underwent such testing. A significant reduction in the number of consultations requested specifically for ECG abnormalities was observed in the POST group (28.6% versus 43.6% in the PRE group; P < 0.0001). Of these consultations referred for ECG abnormalities, the POST group had a greater number that were determined to be abnormal and had fewer noted to have either no changes from previous ECGs or changes resulting from poor lead placement (Table 6). No differences in the number of cases canceled by the cardiologist were noted in the PRE and POST groups (n = 21 [2.3%] and n = 4 [1.4%], respectively; P = 0.47). A total of 20 and 2 patients had their surgery postponed because of the concerns of the cardiologist in the PRE and POST groups, respectively. Eleven patients in the PRE group and 2 in the POST group ultimately had their surgical procedures performed. An additional 31 and 16 patients in the PRE and POST groups, respectively, had their surgery canceled for other reasons (e.g., presence of metastatic disease, infection, patient cancellation of procedure, and so on).
No significant differences were noted in the number and type of postoperative complications experienced by patients who had been evaluated by the cardiologist (Table 7). The overall postoperative complication rate was 18% in both groups, with surgical complications the most prevalent; without consideration of the surgical complications, 13.6% and 14.3% of the consulted patients in the PRE and POST groups, respectively, experienced postoperative morbidity. Two patients in each group had perioperative myocardial infarctions. In the PRE group, stable known or no cardiac disease was noted before surgery in these patients, whereas in the POST group, both patients were thought to be high risk but stable before surgery.
|
Discussion
|
|---|
A growing emphasis in preoperative risk assessment is the provision of high-quality, cost-effective care with the elimination of unnecessary tests and consultations. Cardiology consultations are frequently overused when measured by the percentage that results in no change in perioperative management or outcome (5). In part, this small yield stems from the number and diversity of health care providers ordering the consultations, including nurse practitioners, internists, surgeons, and anesthesiologists, resulting in differences in consultation practices and expertise in preoperative assessment (5). In addition, the number and diversity of cardiologists, even within a single institution, who fulfill consultations create different perspectives on the purposes of the consult (5), the assessment of cardiac risk (6), and the perceived implications of various anesthetic and surgical interventions, procedures, and providers.
Despite a reduction in these confounding factors within our PATC setting, where all consultations are requested by an attending anesthesiologist and fulfilled by a single cardiologist, we anecdotally observed that few completed consultations required additional cardiologist-requested testing, resulted in a new or unstable cardiac diagnosis, or altered perioperative management. In addition, many of the assessments appeared to conclude that the patient was at low cardiac risk. As a result, we speculated that there was an oversubscription for consultations, rather than an overly conservative approach by the cardiologist. Therefore, we implemented significant procedural, educational, and staffing changes within our PATC, focusing on perioperative cardiac assessment. This study was initiated to evaluate the effectiveness of those changes, with the hypotheses that the overall number of consultations would be reduced, the relative testing necessary to evaluate the cardiac status would be increased, and the overall outcome of suspected or known cardiac patients undergoing noncardiac surgeries would at least be maintained, if not improved.
The results of this study support our hypotheses and the value of our interventions. Overall, a 60% reduction in the total number of consultations performed was noted when comparing the three-year study periods comprising the PRE and POST groups. Of note, this reduction was produced despite an increase in the surgical CMI, which indicated that a larger number of patients with comorbidities were being seen in the POST group. In addition, although a similar incidence of consults resulted in the detection of or alteration in cardiac disease, twice as many patients in the POST group (22.7% versus 49.6%; P < 0.0001 in the PRE and POST groups, respectively) required additional investigational testing (exercise tolerance test, echocardiogram, and so on) to arrive at this conclusion. Finally, although the incidence of postoperative complications in patients for whom a consultation was considered but decided against could not be assessed through the available data, the complications in those patients for whom a cardiac consultation was performed did not significantly change in the POST group (Table 7). This may suggest that when a more appropriate threshold for cardiac consultations is established, resulting in patients with potentially higher cardiac risk being seen, improvements in cardiac evaluation, optimization, and management allow for similarly small rates of complications to be observed. Overall, these findings suggest that a more appropriate use of cardiology consultations was made.
We believe that the implementation of three changes within our PATC between the PRE and POST groups was either directly or indirectly responsible for our results. First, more stringent procedural and consultation algorithms were established. Procedural changes included scheduling patients for their PATC visit a greater number of days before their surgical procedure. This change allowed more time to obtain and compare present information with past medical and cardiac-testing records. Second, a formal educational program that was focused on cardiac risk assessment and ECG interpretation was established for both attending and resident physicians. This program included management algorithms for cardiac and ECG anomalies, such as ECGs produced by respiratory variation (9) or incorrect lead placement (10). As a result, a significant reduction in the total number and frequency of consultations requested because of ECG abnormalities decreased from 43.6% to 28.5% (P = 0.0001). In the PRE and POST groups, ECGs found to be due solely to inappropriate lead placement were reduced from 36 to two patients, and dramatic reductions in the number of ECGs ultimately unchanged from previous tracings were observed. When consultations were performed, more individuals with symptoms suggestive of cardiac disease (n = 144, 16%; n = 58, 21%) or alterations in known cardiac disease (n = 183, 20%; n = 75, 27%) were noted in the POST group. These findings underscore the value of and need for educational programs specific to preoperative assessment and evaluation. Previously we reported that although 97% of residency training programs viewed competency in preoperative assessment as an important skill, only 43% had a formal curriculum in this area (11).
Finally, alterations in physician and ancillary staffing were made. Whereas the attending anesthesiology staffing in our PATC setting was previously assigned randomly to any available staff member, the POST group staffing was limited to those with special interests or education in preoperative care. This potentially reduced the variation in the threshold for and quality of the consultations. Nationwide, such alterations in PATC staffing could potentially make a significant effect; in our survey of North American residency programs with PATCs, <10% of their staff had any interest or expertise in preoperative assessment (11). Ancillary staffing improvements potentially reduced consultations as well, by providing a mechanism for previous medical records, ECGs, and verbal or written communication with primary care internists or cardiologists to be obtained for comparison with current data before the anticipated surgical procedure.
Although we attribute our findings to the changes implemented, we recognize that there may be some limitations, as with any nonrandomized, observational study. First, differences in patient demographics or the surgical case mix (e.g., patients with fewer comorbidities or undergoing less invasive surgery) in the POST group may have favored fewer cardiac consultations. However, the surgical CMI increased in all surgical services, and consultations were more frequent in some of the more invasive surgical subspecialties (i.e., thoracic surgery). In addition, more individuals in the POST group were scheduled for admission after surgery, despite a shift in hospital philosophy encouraging ambulatory surgery, reflecting more invasive surgical interventions. These alterations should have resulted in an increase in cardiac consultations.
Second, the ordering of cardiac consultations was still left to the discretion of the attending anesthesiologist in the POST group; thus, improvements in the consultation yield may have been the result of using anesthesiologists with more expertise and comfort in assessing cardiac risk. Yet this idea supports the central tenet of our article: that improvements in consultation yield can be produced by reducing the number of individuals ordering the consultations, educating those individuals in cardiac risk assessment and ECG analysis, and providing algorithms for stratifying risk. Of note, a truly randomized trial to assess the value of cardiac consultations would be difficult to conduct, because patients whom anesthesiologists believed needed cardiac consultations would be required to be placed into a group that would not receive one.
Third, the assessment biases of a single cardiologist may have influenced the results. Although individual consultant biases will always be reflected, we believe that our results remain valid and applicable to other environments. Our confidence is based on the use of a cardiologist who has been the primary preoperative consultant for longer than 25 years, was not on the "learning" portion of his understanding of anesthetic and surgical implications, and used cardiac assessment modalities that were well established. Thus, although our consultant’s biases were certainly expressed, less variation in these assessments would be expected than if a number of independent consultants were used. In addition, it is unlikely that our cardiologist significantly altered his practice between the two study periods, especially because no new cardiac testing modalities were introduced during the study period that could have altered risk stratification. Moreover, our cardiologist was not made aware of the deliberate attempt to decrease consultations originating in our preoperative testing clinic. Although it is not known whether the cardiologist’s recommendations ultimately influenced the anesthetic management, this measure is difficult to assess even in prospective studies; previous studies have indicated that compliance with cardiology consultant recommendations is poor or incomplete (12).
In summary, a significant number of preoperative cardiology consultations do not result in alterations in perioperative testing or management. The thoughtful use of PATCs, with special attention to consultation procedures, education, and staffing, can dramatically reduce the number of and improve the yield of cardiology consultations. These alterations ultimately add to the delivery of the high-quality, cost-effective care that is so important to the safe assessment and delivery of surgical anesthesia in the new millennium.
|
Acknowledgments
|
|---|
Supported by the Department of Anesthesiology, Perioperative and Pain Medicine.
|
Footnotes
|
|---|
Presented in part at the annual meeting of the American Society of Anesthesiologists, Dallas, TX, October 10, 1999.
|
References
|
|---|
- Duetschman CS, Traber KB. Evolution of anesthesiology [editorial]. Anesthesiology 1996; 85: 1–3.[ISI][Medline]
- Mangano DT, Goldman L. Preoperative assessment of patients with known or suspected coronary disease. N Engl J Med 1995; 333: 1750–6.[Free Full Text]
- Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977; 297: 845–50.[Abstract]
- Executive summary of the ACC/AHA task force report: guidelines for perioperative cardiovascular evaluation for noncardiac surgery. Anesth Analg 1996; 82: 854–60.
- Katz RI, Barnhart JM, Ho G, et al. A survey on the intended purposes and perceived utility of preoperative cardiology consultations. Anesth Analg 1998; 87: 830–6.[Abstract/Free Full Text]
- Kleinman B, Czinn E, Shah K, et al. The value to the anesthesia-surgical care team of the preoperative cardiac consultation. J Cardiothorac Anesth 1989; 3: 670–81.
- Rudd P, Siegler M, Byyny RL. Preoperative diabetic consultations: a plea for improved training. J Med Educ 1978; 53: 590–6.[ISI][Medline]
- Dudley JC, Brandenburg JA, Hartley LH, et al. Last-minute preoperative cardiology consultations: epidemiology and impact. Am Heart J 1996; 131: 245–9.[ISI][Medline]
- Mudge GH Jr. The resting electrocardiogram. In: Cohn PF, Wynne J, eds. Diagnostic methods in clinical cardiology. Boston: Little, Brown, 1982: 5–28.
- Peberdy M, Ornato JP. Recognition of electrocardiographic lead misplacements. Am J Emerg Med 1993; 11: 403–5.[ISI][Medline]
- Tsen LC, Segal S, Pothier M, Bader AM. A survey of residency training in preoperative evaluation. Anesthesiology 2000; 93: 1134–7.[Medline]
- Lee TH, Goldman L. Perioperative consultation: the science and the art. J Cardiothorac Anesth 1989; 3: 679–81.[Medline]
Accepted for publication August 14, 2002.
This article has been cited by other articles:
|
|
|
|
 
L. R. Ferrari
Preoperative Evaluation of Pediatric Surgical Patients with Multisystem Considerations
Anesth. Analg.,
October 1, 2004;
99(4):
1058 - 1069.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. L. Hepner, A. M. Bader, S. Hurwitz, M. Gustafson, and L. C. Tsen
Patient Satisfaction with Preoperative Assessment in a Preoperative Assessment Testing Clinic
Anesth. Analg.,
April 1, 2004;
98(4):
1099 - 1105.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
