- Email: [email protected]
Preoperative and postoperative medical evaluation of surgical patients
Preoperative and Postoperative Medical Evaluation of Surgical Patients Robert Gluck, MD, Eric Mu~oz, MD, and Leslie Wise, MD, New Hyde Park, New York
Surgeons continually strive for improvement in the quality of care in the delivery of surgical services. One way to help achieve this goal has been for surgeons to obtain the consultation of their colleagues, especially internists, for the preoperative and postoperative management of surgical patients. In this setting, the internist's role should be to augment the surgeon in the prevention and management of problems contributing to surgical morbidity and mortality [I]. Preoperative and postoperative medical evaluation falls into two broad categories. One category is the general medical preoperative evaluation, also known as medical clearance. In low-risk patients, the surgeon may have no particular reason for wanting a medical evaluation but nonetheless, because of referral dynamics, hospital policy, or medicolegal concerns, asks the internist to evaluate the patient. The second broad category relates to higher risk patients, when the surgeon may feel uncomfortable managing certain medical diseases. In both of these categories of surgical patients medical problems may develop postoperatively, and the internist then acts as a consultant to interpret the more complex medical problems. Many of these postoperative problems are cardiopulmonary, the most common causes of postoperative mortality, and the ones with which the internist may have greater expertise [2]. There are a number of payment changes for hospital and physician care thay may affect the practice of surgery and the use of medical consultation. Hospital inpatient care is undergoing a fundamental change in the payment of Medicare patients. Hospitals are paid a set price according to each of 468 diagnosis-related groups (DRGs) [3]. These changes for hospital care have introduced an economic risk into hospital reimbursement and have decreased lengths of stay and the use of ancillary services. In addition, the Health Care Financing Administration is investigating physician DRG payment systems [4], Physician DRGs may pay a set amount by diagnosis, which may then be divided among all From the Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York and the State Universityof New York at Stony Brook, Stony Brook, New York. Requests for reprints should be addressed to Eric MuSoz, MD, Research Division, Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York 11042.
730
physicians caring for the patient. It is likely that physician DRGs will provide financial disincentives to obtain consultative services. In light of this rapidly changing health care reimbursement environment, the purpose of the present study was to analyze the cost-benefit of routine preoperative and postoperative medical evaluation of surgical patients at a large teaching hospital.
Material a~d Methods The Long Island Jewish Medical Center is an 805 bed voluntary teaching hospital in suburban New York. In addition to teaching programs in all medical and surgical specialties, it has a recovery room and surgical intensive care unit, which are jointly staffed by the departments of surgery and anesthesiology. We reviewed the medical records of 70 patients over the age of 40 (mean age 64 4- 14 years) operated on at the Center between January and June 1984. Patients were selected by the use of a random computer model using medical record numbers, surgical disease category, and American Society of Anesthesiologists (ASA) risk class. These patients represented four different surgical disease categories as follows: (1) major reconstructive vascular procedures (DRG no. 110 and 111); (2)major small and large bowel procedures (no. 148 and 149); (3) cholecystectomy without common duct exploration (no. 197 and 198); and (4) major head and neck procedures (no. 49). Surgical disease categories were selected within the DRG format to represent common general surgical or vascular procedures. Our hospital regulations require routine medical evaluation of all patients over 40 years of age prior to surgery regardless of surgical risk. Each patient was independentlygraded by us according to surgical risk using the Anesthesiologists Physical Status Scale [5-7] as follows: class 1, a normally healthy individual; class 2, a patient with mild systemic disease; class 3, a patient with severe systemic disease that is not incapacitating; class 4, a patient with an incapacitating systemic disease that is a constant threat to life; and class 5, a moribund patient who is not expected to survive 24 hours with or without operation. These grades were corroborated by review of the anesthesia records on which they are routinely recorded. To simplify the study, patients in classes 4 and 5 (very high risk) and nonelective cases (emergency admissions) were excluded from the study.
The American Journal of Surgery
Medical Evaluation for Surgical Patients
TABLE I
All Preoperative and Postoperative RecommendaUons for American Society of Anesthesiologists Classes 1 and 2 Preoperative
1. 2. 3. 4. 5.
Transfuse blood Repeat urinalysis for microscopic hematuria Use hydralazine instead of methyldopa (Aldomel| for allergy Prophylactic antibiotics Subacute bacterial endocarditis prophylaxis
The preoperative medical evaluation and follow-up for the first postoperative week were reviewed by us for each patient. The following indices were recorded: (1) total number of preoperative recommendations by the internist; (2) number of preoperative recommendations considered significant; (3) number of patient visits (consultations) by the internist during the first postoperative week; (4) total number of postoperative recommendations by the internist; and (5) number of postoperative recommendations considered significant. Recommendations were also reviewed independently by two members of the attending surgical staff. A recommendation was considered significant if it was outside the normal purview and experience of the surgical staff and support system at our institution. By way of example, recommendations regarding routine steroid and insulin adjustments, antibiotic prophylaxis, fluid management, nutritional support, and fever work-up were not considered significant. Where multiple recommendations were made at one time or the same recommendation was repeated, this was regarded as one recommendation. A complete list of all recommendations made by the medical consultants for ASA classes 1 and 2 is shown in Table I. Results were tabulated for each surgical risk group and for the total population. Costs of medical consultation were based on average fees for the study period (1984) at our institution as follows: $100 per preoperative medical evaluation and $60 per postoperative medical consultation (visit). Patients in the study population were cared for by a total of 27 different attending surgeons and 25 different internists. The Student's t test was used to compare statistical significance between groups. Results
The number of preoperative and postoperative recommendations according to ASA class are shown in Table II. In ASA class 1 (12 patients), there were no preoperative or postoperative recommendations. In ASA class 2 (27 patients), there were three preoperative recommendations made in three patients (11 percent of patients), none of which were significant. Postoperatively, there were five recommendations in five patients (19 percent of patients), none of which were significant. The total number of pre-
Volume 155, June 1988
Postoperative
1. 2. 3. 4. 5. 6. 7. 8. 9.
Check laboratory test (complete blood count, urinalysis) Continue preoperative medications Monitor intake & output Discontinue Foley catheter Workup of confusion (order laboratory tests, CAT scan) Insulin and metoprolol tartrate (Lopressor~) dosage Administer antacids Work-up fever Order hepatitis screen for elevated liver enzymes
operative and postoperative recommendations for class 2 was eight in eight different patients (30 percent of patients), none of which were significant. In class 3 (31 patients), there were 12 preoperative recommendations in 10 different patients (32 percent of patients), 3 recommendations of which were significant; that is, 25 percent of recommendations in three patients or 10 percent of patients in the class had significant recommendations. Postoperatively, there were 22 recommendations in 20 patients (65 percent of patients in the class), 7 recommendations of which were significant (32 percent of recommendations in 19 percent of the patients in the class). The total number of preoperative and postoperative recommendations for class 3 was 34 in 30 different patients (97 percent of patients), 10 recommendations of which were significant (33 percent of recommendations in 9 different patients or 27 percent of the patients in the class). Preoperative and postoperative total and significant recommendations were statistically significantly different for classes I and II versus class III as demonstrated in Table II. The aggregate results for all 70 patients regarding preoperative and postoperative recommendations were as follows: 15 preoperative recommendations in 13 different patients (19 percent of patients), 3 recommendations of which were significant (20 percent of recommendations) in 3 different patients (4 percent of the patients); 27 postoperative recommendations in 25 different patients (36 percent of patients), 7 recommendations of which were significant (26 percent of recommendations) in 6 different patients (9 percent of patients); and a total of 42 recommendations in 38 different patients (54 percent of patients), 10 of which were significant (24 percent of recommendations) in 9 different patients (13 percent of patients). There was a total of six postoperative deaths, of which two were in class 2 patients (minor upper gastrointestinal bleeding not requiring transfusion and atrial fibrillation requiring digitalization) and four were in class 3 patients (atrial fibrillation requiring digitalization; nonsignificant premature
731
Gluck et al
TABLE II
Preoperative and Postoperative Recommendations by American Society of Anesthesiologists Class Preoperative*
Total Glass 1 (n = 12) Class 2 (n = 27) Class 3 (n = 31) Total (n = 70)
R
Pw
0
...
Postoperative t
Significant R82 P
Total
Significant
R
P
...
0
...
0 5
5 (19)
R
P
0
...
0
0 (0)
3
3 (11)
0 (0)
. ..
12
10 (32)
3 (25)
3 (10)
22
20 (65)
7 (32)
6 (19)
15
13 (19)
3 (20)
3 (4)
27
25 (36)
7 (26)
6 (9)
* Total: class I versus II, p <0.73; class I versus III, p <0.0001; class II versus III, p <0.01. Significant: class I versus II, p <1; class I t Total: class I versus II, p <0.02; class I versus III, p <0.0001; class II versus III, p <0.0001). Significant: class I versus II, p <1; class Mean 4- standard deviation. w Parenthetical data in patient columns indicate percentages of patients in each class. 82Parenthetical data in recommendations columns indicate percentage of recommendations which were significant. P = patients; R = recommendations.
TABLE III
Total Medical Consultation Fees According to American Society of Anesthesiologists Cost/ Preop
Class 1 (n = 12) Class 2 (n = 27) Class 3 (n = 31) Total
Postop
Total
Patient
$1,200
$1,368
$2,568
$214
$2,700
$5,670
$8,370
$310
$3,100 $7,000
$8,370 $15,408
$11,470 $22,406
$370 $320
ventricular contractions; atrial premature contractions; and development of tachypnea, tachycardia, and mild renal failure). There was one death on the first postoperative day in a class 3 patient. This death was secondary to disseminated intravascular coagulopathy and subsequent myocardial infarction. The total medical consultation fees for each ASA class are listed in Table III. Comments
The purpose of this study was to assess the costbenefit of routine preoperative and postoperative medical evaluation of surgical patients. Our data showed no significant recommendations made either preoperatively or postoperatively for the two lower risk patient classes we studied (ASA classes 1 and 2). In fact, in these two groups totaling 39 patients, there were only 8 recommendations of any kind. The mean number of postoperative visits by the medical consultant for these groups was 1.9 and 3.5 (classes 1 and 2, respectively). The aggregate costs of physician fees for medical consultations for these two groups was $10,938 ($280 per patient).
732
Some significant recommendations were made, however, in class 3 patients (the medium risk group). The mean number of postoperative visits was 4.5. This is not surprising since this group represented sicker patients, consistent with their resultant higher morbidity and mortality rates, as shown in this and previous studies [8,9]. Since the medical consultants made no significant recommendations for the classes 1 and 2 patients, it is unlikely that they contributed to any sizeable reduction in morbidity or mortality in these groups. They may, however, have contributed to a reduction in morbidity and mortality in class 3 patients, since there were 10 significant recommendations made in this group in 9 different patients (29 percent of patients in the class). The disease categories included in the present study were chosen arbitrarily because they included a wide variety of commonly performed general and vascular surgical procedures of various risk levels. Though subjective, the ASA physical status scale demonstrates good agreement among physicians on the grading of risk class [6]. Moreover, there is a strong correlation between the ASA physical status and operative mortality (Table IV) [7-9]. According to Goldman [10], when the ASA scale was utilized postoperatively, it was the best available predictor of noncardiac deaths and a fair predictor of cardiac deaths. As a first step, we analyzed ASA classes 1, 2, and 3 patients. Further study will be required regarding the value of medical evaluation for the more seriously ill patient (ASA classes 4 and 5) and emergency patients. Surgeons must be selective in their use of resources. They must avoid extensive redundancy in patient management and the temptation to pass the buck, either in a medical or medicolegal sense. Without this restraint, they will, at the very least, incur unnecessary expense in this age of shrinking
The American Journal of Surgery
Medical Evaluation for Surgical Patients
TABLE IV Overall Total R 0
Total Significant P
...
R
P
Postoperative Visits (n)$
0
...
1.9 4- 1.4
0
...
3.5 4- 2.2 4.5 4- 3.!
8
8 (30)
34
30 (97)
10 (33)
9 (29)
42
38 (54)
10 (24)
9 (13)
versus III, p <0.01; class II versus IIIi p <0.01. I versus III, p <0.006; class II versus III, p <0.006).
health care dollars. Considering the yearly number of operations in low- and medium-risk patients and the cost of medical consultation, the aggregate cost nationwide may be very high. Assuming 10 million annual inpatient procedures of low risk (ASA classes 1 and 2) with a per patient cost of $280 for medical consultations, as at our hospital, $2.8 billion would be generated for medical consultant fees. The present study examined the cost-benefit of medicalevaluation for low- and medium-risk surgical patients. Physician DRGs would provide the most immediate disincentive for the use of low-benefit medical consultation. Under physician DRGs, all services performed by physicians would be combined in a single bill, with a single payment made. This type of payment system would encourage surgeons to cut back on marginally necessary tests or consultations that generate physician's fees, since their costs must come out of the fixed case payment. There are still many unanswered questions concerning physician DRGs, such as who receives the payment, and how it is divided. Although this study has demonstrated a limited benefit from medical evaluation, the dominant reasons for medical evaluation may relate to matters such as referral dynamics, hospital regulations, and mediColegal liability. Other justifications for such consultation in low-risk patients are that preoperative medical evaluation may lead to more efficient and better informed postoperative care.That is, if a problem arises postoperatively, the internist will not then be called to see the patient for the first time. This argument appears faulty for a number of reasons. First, morbidity and mortality rates in lowrisk patients are low (Table IV). Second, as shown in the present study, in the groups of surgical patients at low risk, no significant recommendations were made. Third, there appears to be no reason why surgeons themselves cannot deal effectively with most of the problems that arise. During 5 years of surgical residency training, surgeons are expected to Volume 155, June 1988
Mortality Rates According to American Standards Association Physica! Status (literature review) ASA Physical Status 1 2 3
48 hour mortality rate (%) for elective surgery [8] 6 week mortality rate (%) for surgical procedures [9] Elective low-risk surgery (n = 366,300) Elective medium-risk surgery (n = 426,000) Elective high-.risk surgery (n =61,700)
0.07
0.24
1.4
0.18
1.4
5.4
0.03
0.31
1.9
0.02
1.7
7.2
2.2
5.4
14.6
become proficient not only in operative techniques, but in the preoperative and postoperative care of surgical patients. PrOficiency in this area is one of the prerequisites for board certification. The Committee on Continuing Education of the American College of Surgeons emphasizes proficiency in this area as part of residency training. In contrast, internists are not specifically trained in preoperative and postoperative care. Adler [11], an internist writing in Medical Evaluation of the Surgical Patient, acknowledges t h a t "many internists feel insecure When called on to evaluate the preoperative patient because this important aspect of medicine is not formally taught during medical training." Another possible motive for preoperative medical evaluation in low-r!sk patients is that it may offer a unique opportunity for periodi c screening of the healthy patient [12]. However, review of the sensitivity, specificity, and yield of many routine screen: ing tests and procedures has shown them to be minimally cost-effective [13-15]. Such data notwithstanding,the surgeon has the option of ordering the necessary tests and procedures as readily as the internist. If help is needed in the interpretation of test results, this can be obtained. Also, at many hospitals such as ours, a system of checks and balances already exists in the form of housestaff and anesthesiologists who routinely evaluate patients preoperatively and~postoperatively. Ultimately, surgeons must individualize their needs for preoperative and postoperative medical evaluation based on their experience, the particula r hospital environment, t h e particular patient population, and the changing incentives for same byway of changes in health care reimbursement: Summary
The purpose of this study was to analyze the costbenefit of routine preoperative and postoperative medical evaluation of surgical patients. We analyzed the records of 70 patients who underwent selected common general and vascular surgical pro733
Gluck et al
cedures according to their surgical risk, and assessed the potential change in outcome from medical evaluation. Our findings suggest t h a t for low-risk surgical patients undergoing these procedures, medical evaluation is unlikely to affect outcome and, thus, may be unnecessary. There appears to be some benefit, however, for medium-risk surgical patients. Although we do not suggest the abolishment of al ! preoperative and postoperative medical evaluations, we do recommend the much more selective use of this valuable resource. Our data suggest that significant improvement in efficiency for many hospitals may be possible by omitting routine preoperative and postoperative medical evaluation for certain patients and by obtaining medical evaluation only in selected medium and higher risk surgical patients (ASA class 3 and above). References 1. Feigal DW, Blaisdell FW. The estimation of surgical risk. Med Clin North Am 1979; 63:1131-44. 2. Committee on Records. American Society of Anesthesiologists Newsletter 1971; 27: 4.
734
3. Iglehart JD. Medicare begins prospective payment of hospitals. N Engl J Med 1983; 308: 1428-32. 4. MitchellJB.PhysiciansDRGs.NEnglJMed 1985;3!3:670-5. 5. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA 1961; 178: 26t-6. 6. Owens WD, Felts JA, Spitnage EL. ASA physical status classification: a study of consistency of ratings. Anesthesiology 1978; 49: 239-43. 7. Del Guercio LRM, Cohn JD. Monitoring operative risk in the elderly. JAMA 1980; 243: 1350-5. 8. VacanU CJ, Van Houten RJ, Hill RC. A statistical analysis of the relationship of physical status to postoperative mortality in 68,388 cases. Anesth Analg 1970; 49: 564-6. 9. Bunker JP, Forrest WH, Mosteller F, et al, eds. The National halothane study: a study of the possible association between ha!othane and postoperative hepatic necrosis, Bethesda: National Institute of Health, National Institute of General Medical Sciences, 1969, 10. Goldman L. Letter. N Engl J Med 1980; 298: 340. 11. Adler AG, ed. Medical evaluation of the surgical patient. Philadelphia: WB Saunders, 1985, 12. Robbins JA, Mushlin AL. Preoperative evaluation of the healthy patient. Med Clin North Am 1979; 63:1145-56. 13. Goldman L. Which tests are necessary for "healthy" patients? Consultants 1984; 24: 331. 14. RobbinsJ, RoseS. Partial thromboplastin time screening test. Ann intern Med 1979; 90: 796-7. 15. Rucker L, Frye EB, Staten MA. Usefulness of screening chart roentgenograms in preoperative patients. JAMA 1983; 250:3209-11,
The American Journal of Surgery
Surgeons continually strive for improvement in the quality of care in the delivery of surgical services. One way to help achieve this goal has been for surgeons to obtain the consultation of their colleagues, especially internists, for the preoperative and postoperative management of surgical patients. In this setting, the internist's role should be to augment the surgeon in the prevention and management of problems contributing to surgical morbidity and mortality [I]. Preoperative and postoperative medical evaluation falls into two broad categories. One category is the general medical preoperative evaluation, also known as medical clearance. In low-risk patients, the surgeon may have no particular reason for wanting a medical evaluation but nonetheless, because of referral dynamics, hospital policy, or medicolegal concerns, asks the internist to evaluate the patient. The second broad category relates to higher risk patients, when the surgeon may feel uncomfortable managing certain medical diseases. In both of these categories of surgical patients medical problems may develop postoperatively, and the internist then acts as a consultant to interpret the more complex medical problems. Many of these postoperative problems are cardiopulmonary, the most common causes of postoperative mortality, and the ones with which the internist may have greater expertise [2]. There are a number of payment changes for hospital and physician care thay may affect the practice of surgery and the use of medical consultation. Hospital inpatient care is undergoing a fundamental change in the payment of Medicare patients. Hospitals are paid a set price according to each of 468 diagnosis-related groups (DRGs) [3]. These changes for hospital care have introduced an economic risk into hospital reimbursement and have decreased lengths of stay and the use of ancillary services. In addition, the Health Care Financing Administration is investigating physician DRG payment systems [4], Physician DRGs may pay a set amount by diagnosis, which may then be divided among all From the Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York and the State Universityof New York at Stony Brook, Stony Brook, New York. Requests for reprints should be addressed to Eric MuSoz, MD, Research Division, Department of Surgery, Long Island Jewish Medical Center, New Hyde Park, New York 11042.
730
physicians caring for the patient. It is likely that physician DRGs will provide financial disincentives to obtain consultative services. In light of this rapidly changing health care reimbursement environment, the purpose of the present study was to analyze the cost-benefit of routine preoperative and postoperative medical evaluation of surgical patients at a large teaching hospital.
Material a~d Methods The Long Island Jewish Medical Center is an 805 bed voluntary teaching hospital in suburban New York. In addition to teaching programs in all medical and surgical specialties, it has a recovery room and surgical intensive care unit, which are jointly staffed by the departments of surgery and anesthesiology. We reviewed the medical records of 70 patients over the age of 40 (mean age 64 4- 14 years) operated on at the Center between January and June 1984. Patients were selected by the use of a random computer model using medical record numbers, surgical disease category, and American Society of Anesthesiologists (ASA) risk class. These patients represented four different surgical disease categories as follows: (1) major reconstructive vascular procedures (DRG no. 110 and 111); (2)major small and large bowel procedures (no. 148 and 149); (3) cholecystectomy without common duct exploration (no. 197 and 198); and (4) major head and neck procedures (no. 49). Surgical disease categories were selected within the DRG format to represent common general surgical or vascular procedures. Our hospital regulations require routine medical evaluation of all patients over 40 years of age prior to surgery regardless of surgical risk. Each patient was independentlygraded by us according to surgical risk using the Anesthesiologists Physical Status Scale [5-7] as follows: class 1, a normally healthy individual; class 2, a patient with mild systemic disease; class 3, a patient with severe systemic disease that is not incapacitating; class 4, a patient with an incapacitating systemic disease that is a constant threat to life; and class 5, a moribund patient who is not expected to survive 24 hours with or without operation. These grades were corroborated by review of the anesthesia records on which they are routinely recorded. To simplify the study, patients in classes 4 and 5 (very high risk) and nonelective cases (emergency admissions) were excluded from the study.
The American Journal of Surgery
Medical Evaluation for Surgical Patients
TABLE I
All Preoperative and Postoperative RecommendaUons for American Society of Anesthesiologists Classes 1 and 2 Preoperative
1. 2. 3. 4. 5.
Transfuse blood Repeat urinalysis for microscopic hematuria Use hydralazine instead of methyldopa (Aldomel| for allergy Prophylactic antibiotics Subacute bacterial endocarditis prophylaxis
The preoperative medical evaluation and follow-up for the first postoperative week were reviewed by us for each patient. The following indices were recorded: (1) total number of preoperative recommendations by the internist; (2) number of preoperative recommendations considered significant; (3) number of patient visits (consultations) by the internist during the first postoperative week; (4) total number of postoperative recommendations by the internist; and (5) number of postoperative recommendations considered significant. Recommendations were also reviewed independently by two members of the attending surgical staff. A recommendation was considered significant if it was outside the normal purview and experience of the surgical staff and support system at our institution. By way of example, recommendations regarding routine steroid and insulin adjustments, antibiotic prophylaxis, fluid management, nutritional support, and fever work-up were not considered significant. Where multiple recommendations were made at one time or the same recommendation was repeated, this was regarded as one recommendation. A complete list of all recommendations made by the medical consultants for ASA classes 1 and 2 is shown in Table I. Results were tabulated for each surgical risk group and for the total population. Costs of medical consultation were based on average fees for the study period (1984) at our institution as follows: $100 per preoperative medical evaluation and $60 per postoperative medical consultation (visit). Patients in the study population were cared for by a total of 27 different attending surgeons and 25 different internists. The Student's t test was used to compare statistical significance between groups. Results
The number of preoperative and postoperative recommendations according to ASA class are shown in Table II. In ASA class 1 (12 patients), there were no preoperative or postoperative recommendations. In ASA class 2 (27 patients), there were three preoperative recommendations made in three patients (11 percent of patients), none of which were significant. Postoperatively, there were five recommendations in five patients (19 percent of patients), none of which were significant. The total number of pre-
Volume 155, June 1988
Postoperative
1. 2. 3. 4. 5. 6. 7. 8. 9.
Check laboratory test (complete blood count, urinalysis) Continue preoperative medications Monitor intake & output Discontinue Foley catheter Workup of confusion (order laboratory tests, CAT scan) Insulin and metoprolol tartrate (Lopressor~) dosage Administer antacids Work-up fever Order hepatitis screen for elevated liver enzymes
operative and postoperative recommendations for class 2 was eight in eight different patients (30 percent of patients), none of which were significant. In class 3 (31 patients), there were 12 preoperative recommendations in 10 different patients (32 percent of patients), 3 recommendations of which were significant; that is, 25 percent of recommendations in three patients or 10 percent of patients in the class had significant recommendations. Postoperatively, there were 22 recommendations in 20 patients (65 percent of patients in the class), 7 recommendations of which were significant (32 percent of recommendations in 19 percent of the patients in the class). The total number of preoperative and postoperative recommendations for class 3 was 34 in 30 different patients (97 percent of patients), 10 recommendations of which were significant (33 percent of recommendations in 9 different patients or 27 percent of the patients in the class). Preoperative and postoperative total and significant recommendations were statistically significantly different for classes I and II versus class III as demonstrated in Table II. The aggregate results for all 70 patients regarding preoperative and postoperative recommendations were as follows: 15 preoperative recommendations in 13 different patients (19 percent of patients), 3 recommendations of which were significant (20 percent of recommendations) in 3 different patients (4 percent of the patients); 27 postoperative recommendations in 25 different patients (36 percent of patients), 7 recommendations of which were significant (26 percent of recommendations) in 6 different patients (9 percent of patients); and a total of 42 recommendations in 38 different patients (54 percent of patients), 10 of which were significant (24 percent of recommendations) in 9 different patients (13 percent of patients). There was a total of six postoperative deaths, of which two were in class 2 patients (minor upper gastrointestinal bleeding not requiring transfusion and atrial fibrillation requiring digitalization) and four were in class 3 patients (atrial fibrillation requiring digitalization; nonsignificant premature
731
Gluck et al
TABLE II
Preoperative and Postoperative Recommendations by American Society of Anesthesiologists Class Preoperative*
Total Glass 1 (n = 12) Class 2 (n = 27) Class 3 (n = 31) Total (n = 70)
R
Pw
0
...
Postoperative t
Significant R82 P
Total
Significant
R
P
...
0
...
0 5
5 (19)
R
P
0
...
0
0 (0)
3
3 (11)
0 (0)
. ..
12
10 (32)
3 (25)
3 (10)
22
20 (65)
7 (32)
6 (19)
15
13 (19)
3 (20)
3 (4)
27
25 (36)
7 (26)
6 (9)
* Total: class I versus II, p <0.73; class I versus III, p <0.0001; class II versus III, p <0.01. Significant: class I versus II, p <1; class I t Total: class I versus II, p <0.02; class I versus III, p <0.0001; class II versus III, p <0.0001). Significant: class I versus II, p <1; class Mean 4- standard deviation. w Parenthetical data in patient columns indicate percentages of patients in each class. 82Parenthetical data in recommendations columns indicate percentage of recommendations which were significant. P = patients; R = recommendations.
TABLE III
Total Medical Consultation Fees According to American Society of Anesthesiologists Cost/ Preop
Class 1 (n = 12) Class 2 (n = 27) Class 3 (n = 31) Total
Postop
Total
Patient
$1,200
$1,368
$2,568
$214
$2,700
$5,670
$8,370
$310
$3,100 $7,000
$8,370 $15,408
$11,470 $22,406
$370 $320
ventricular contractions; atrial premature contractions; and development of tachypnea, tachycardia, and mild renal failure). There was one death on the first postoperative day in a class 3 patient. This death was secondary to disseminated intravascular coagulopathy and subsequent myocardial infarction. The total medical consultation fees for each ASA class are listed in Table III. Comments
The purpose of this study was to assess the costbenefit of routine preoperative and postoperative medical evaluation of surgical patients. Our data showed no significant recommendations made either preoperatively or postoperatively for the two lower risk patient classes we studied (ASA classes 1 and 2). In fact, in these two groups totaling 39 patients, there were only 8 recommendations of any kind. The mean number of postoperative visits by the medical consultant for these groups was 1.9 and 3.5 (classes 1 and 2, respectively). The aggregate costs of physician fees for medical consultations for these two groups was $10,938 ($280 per patient).
732
Some significant recommendations were made, however, in class 3 patients (the medium risk group). The mean number of postoperative visits was 4.5. This is not surprising since this group represented sicker patients, consistent with their resultant higher morbidity and mortality rates, as shown in this and previous studies [8,9]. Since the medical consultants made no significant recommendations for the classes 1 and 2 patients, it is unlikely that they contributed to any sizeable reduction in morbidity or mortality in these groups. They may, however, have contributed to a reduction in morbidity and mortality in class 3 patients, since there were 10 significant recommendations made in this group in 9 different patients (29 percent of patients in the class). The disease categories included in the present study were chosen arbitrarily because they included a wide variety of commonly performed general and vascular surgical procedures of various risk levels. Though subjective, the ASA physical status scale demonstrates good agreement among physicians on the grading of risk class [6]. Moreover, there is a strong correlation between the ASA physical status and operative mortality (Table IV) [7-9]. According to Goldman [10], when the ASA scale was utilized postoperatively, it was the best available predictor of noncardiac deaths and a fair predictor of cardiac deaths. As a first step, we analyzed ASA classes 1, 2, and 3 patients. Further study will be required regarding the value of medical evaluation for the more seriously ill patient (ASA classes 4 and 5) and emergency patients. Surgeons must be selective in their use of resources. They must avoid extensive redundancy in patient management and the temptation to pass the buck, either in a medical or medicolegal sense. Without this restraint, they will, at the very least, incur unnecessary expense in this age of shrinking
The American Journal of Surgery
Medical Evaluation for Surgical Patients
TABLE IV Overall Total R 0
Total Significant P
...
R
P
Postoperative Visits (n)$
0
...
1.9 4- 1.4
0
...
3.5 4- 2.2 4.5 4- 3.!
8
8 (30)
34
30 (97)
10 (33)
9 (29)
42
38 (54)
10 (24)
9 (13)
versus III, p <0.01; class II versus IIIi p <0.01. I versus III, p <0.006; class II versus III, p <0.006).
health care dollars. Considering the yearly number of operations in low- and medium-risk patients and the cost of medical consultation, the aggregate cost nationwide may be very high. Assuming 10 million annual inpatient procedures of low risk (ASA classes 1 and 2) with a per patient cost of $280 for medical consultations, as at our hospital, $2.8 billion would be generated for medical consultant fees. The present study examined the cost-benefit of medicalevaluation for low- and medium-risk surgical patients. Physician DRGs would provide the most immediate disincentive for the use of low-benefit medical consultation. Under physician DRGs, all services performed by physicians would be combined in a single bill, with a single payment made. This type of payment system would encourage surgeons to cut back on marginally necessary tests or consultations that generate physician's fees, since their costs must come out of the fixed case payment. There are still many unanswered questions concerning physician DRGs, such as who receives the payment, and how it is divided. Although this study has demonstrated a limited benefit from medical evaluation, the dominant reasons for medical evaluation may relate to matters such as referral dynamics, hospital regulations, and mediColegal liability. Other justifications for such consultation in low-risk patients are that preoperative medical evaluation may lead to more efficient and better informed postoperative care.That is, if a problem arises postoperatively, the internist will not then be called to see the patient for the first time. This argument appears faulty for a number of reasons. First, morbidity and mortality rates in lowrisk patients are low (Table IV). Second, as shown in the present study, in the groups of surgical patients at low risk, no significant recommendations were made. Third, there appears to be no reason why surgeons themselves cannot deal effectively with most of the problems that arise. During 5 years of surgical residency training, surgeons are expected to Volume 155, June 1988
Mortality Rates According to American Standards Association Physica! Status (literature review) ASA Physical Status 1 2 3
48 hour mortality rate (%) for elective surgery [8] 6 week mortality rate (%) for surgical procedures [9] Elective low-risk surgery (n = 366,300) Elective medium-risk surgery (n = 426,000) Elective high-.risk surgery (n =61,700)
0.07
0.24
1.4
0.18
1.4
5.4
0.03
0.31
1.9
0.02
1.7
7.2
2.2
5.4
14.6
become proficient not only in operative techniques, but in the preoperative and postoperative care of surgical patients. PrOficiency in this area is one of the prerequisites for board certification. The Committee on Continuing Education of the American College of Surgeons emphasizes proficiency in this area as part of residency training. In contrast, internists are not specifically trained in preoperative and postoperative care. Adler [11], an internist writing in Medical Evaluation of the Surgical Patient, acknowledges t h a t "many internists feel insecure When called on to evaluate the preoperative patient because this important aspect of medicine is not formally taught during medical training." Another possible motive for preoperative medical evaluation in low-r!sk patients is that it may offer a unique opportunity for periodi c screening of the healthy patient [12]. However, review of the sensitivity, specificity, and yield of many routine screen: ing tests and procedures has shown them to be minimally cost-effective [13-15]. Such data notwithstanding,the surgeon has the option of ordering the necessary tests and procedures as readily as the internist. If help is needed in the interpretation of test results, this can be obtained. Also, at many hospitals such as ours, a system of checks and balances already exists in the form of housestaff and anesthesiologists who routinely evaluate patients preoperatively and~postoperatively. Ultimately, surgeons must individualize their needs for preoperative and postoperative medical evaluation based on their experience, the particula r hospital environment, t h e particular patient population, and the changing incentives for same byway of changes in health care reimbursement: Summary
The purpose of this study was to analyze the costbenefit of routine preoperative and postoperative medical evaluation of surgical patients. We analyzed the records of 70 patients who underwent selected common general and vascular surgical pro733
Gluck et al
cedures according to their surgical risk, and assessed the potential change in outcome from medical evaluation. Our findings suggest t h a t for low-risk surgical patients undergoing these procedures, medical evaluation is unlikely to affect outcome and, thus, may be unnecessary. There appears to be some benefit, however, for medium-risk surgical patients. Although we do not suggest the abolishment of al ! preoperative and postoperative medical evaluations, we do recommend the much more selective use of this valuable resource. Our data suggest that significant improvement in efficiency for many hospitals may be possible by omitting routine preoperative and postoperative medical evaluation for certain patients and by obtaining medical evaluation only in selected medium and higher risk surgical patients (ASA class 3 and above). References 1. Feigal DW, Blaisdell FW. The estimation of surgical risk. Med Clin North Am 1979; 63:1131-44. 2. Committee on Records. American Society of Anesthesiologists Newsletter 1971; 27: 4.
734
3. Iglehart JD. Medicare begins prospective payment of hospitals. N Engl J Med 1983; 308: 1428-32. 4. MitchellJB.PhysiciansDRGs.NEnglJMed 1985;3!3:670-5. 5. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality. JAMA 1961; 178: 26t-6. 6. Owens WD, Felts JA, Spitnage EL. ASA physical status classification: a study of consistency of ratings. Anesthesiology 1978; 49: 239-43. 7. Del Guercio LRM, Cohn JD. Monitoring operative risk in the elderly. JAMA 1980; 243: 1350-5. 8. VacanU CJ, Van Houten RJ, Hill RC. A statistical analysis of the relationship of physical status to postoperative mortality in 68,388 cases. Anesth Analg 1970; 49: 564-6. 9. Bunker JP, Forrest WH, Mosteller F, et al, eds. The National halothane study: a study of the possible association between ha!othane and postoperative hepatic necrosis, Bethesda: National Institute of Health, National Institute of General Medical Sciences, 1969, 10. Goldman L. Letter. N Engl J Med 1980; 298: 340. 11. Adler AG, ed. Medical evaluation of the surgical patient. Philadelphia: WB Saunders, 1985, 12. Robbins JA, Mushlin AL. Preoperative evaluation of the healthy patient. Med Clin North Am 1979; 63:1145-56. 13. Goldman L. Which tests are necessary for "healthy" patients? Consultants 1984; 24: 331. 14. RobbinsJ, RoseS. Partial thromboplastin time screening test. Ann intern Med 1979; 90: 796-7. 15. Rucker L, Frye EB, Staten MA. Usefulness of screening chart roentgenograms in preoperative patients. JAMA 1983; 250:3209-11,
The American Journal of Surgery