Cost of ectopic pregnancy management: surgery versus methotrexate * t

Transcription

1 FERTILITY AND STERILITY Copyright c 1993 The American Fertility Society Printed on acid-free paper in U. S. A. Cost of ectopic pregnancy management: surgery versus methotrexate * t Mitchell D. Creinin, M.D.:j: A. Eugene Washington, M.D., M.Sc.:j: University of California School of Medicine, San Francisco, San Francisco, California Objective: To estimate the potential annual cost savings of methotrexate (MTX) therapy for ectopic pregnancy (EP) at our institution and nationally. Design: Retrospective. Setting: San Francisco General Hospital. Patients: All patients treated for EP in Interventions: Medical records and billing statements were reviewed. Main Outcome Measures: Direct costs were determined from billing statements, and indirect costs were estimated based on published data. Potential annual national cost savings from MTX treatment were then extrapolated using national statistics on the incidence of EP. Results: Fifty EPs were treated with an average direct cost of $7,972 per case. Direct and indirect costs totaled >$500,000. Fifteen (30%) of the cases would have been eligible for MTX treatment with a resultant cost savings of >$160,000. The potential annual national cost savings would be in excess of $280 million. Conclusions: If preliminary findings suggesting the effectiveness of MTX as an alternative therapy for EP are confirmed, appropriate use of MTX will not only reduce morbidity but also the substantial cost. Fertil Steril 1993;60:963-9 Key Words: Ectopic pregnancy, methotrexate, cost From 1970 to 1989, the incidence of ectopic pregnancy (EP) more than tripled; in 1989, 88,400 EPs were reported, accounting for 1.6% of all pregnancies (1). This rise has resulted in an increase in EPrelated sequelae that diminish women's reproductive capability. Additionally, EP accounts for 13% of all pregnancy-related deaths, making it one of the leading causes of maternal mortality (2). During the late 1980s, in an attempt to decrease both morbidity and cost, treatment for EP began to Received April 8, 1993; revised and accepted August 19, * Supported by grant number HS from the Agency for Health Care Policy and Research. t Reprints not available. :j: Department of Obstetrics, Gynecology, and Reproductive Sciences. MEDTEP Research Center on Minority Populations, Institute for Health Policy Studies. focus on laparoscopic modalities. Two prospective randomized trials in the United States documented the safety of laparoscopy compared with laparotomy for the treatment of tubal pregnancy (3, 4). Both studies showed that the two procedures have equal efficacy, similar complication rates, and require the same operative time. There was also less blood loss and a shorter average hospital stay (approximately 1 versus 3 days) with laparoscopy, and shorter recovery time after hospitalization for laparoscopy patients. In the late 1980s, reports of successful use of methotrexate (MTX) as a nonsurgical treatment for EP also began to appear in the literature (5). Initially, treatment involved multidose systemic MTX with leucovorin rescue. Recently, Stovall and co-workers (6) reported a single-dose intramuscular injection protocol equal in efficacy to multidose regimens and with minimal side effects (6). Creinin and Washington Cost of ectopic treatment 963

2 Although the key criteria for evaluation and acceptance of a new technology is improvement of health care with a resultant decrease in morbidity and mortality, the accompanying important economic consequences of the disease and its treatment are important as well. The estimated cost of EP in the United States for 1990 was $1.1 billion, including both hospital charges and disability/lost wages (7). This cost does not include the subsequent morbidity that is incurred as a result of EP. Although the standard of care in 1990 included routine treatment via laparotomy, it has been shown that the hospital cost savings for laparoscopic treatment compared with traditional laparotomy is approximately $1,200 to 1,500 per case (3, 4). Additionally, a report of cost savings from England in 1989 including loss of wages from disability showed the overall health and social service costs oflaparoscopy were half that of laparotomy (8). With MTX emerging as a possible alternative management for selected women with an EP, an analysis of its comparative costs is now indicated. In this article, we examine the costs associated with management of EP in a local hospital and use these estimates to project potential cost savings of MTX treatment of EP. MATERIALS AND METHODS We reviewed the medical records and billing statements of all patients discharged from San Francisco General Hospital from January 1 through December 31,1991, with a diagnosis ofep. Patients were identified using principal diagnosis codes of through (International Classification of Diseases, Ninth Revision, Clinical Modification) (9). Patients were deemed to be "eligible for MTX treatment" if they satisfied the following conditions: [1] they were hemodynamically stable; [2] the size of the EP on ultrasound (US) before the procedure or at the time of surgery was not >3.5 cm in its greatest dimension; [3] the EP was located in the fallopian tube and not ruptured at the time of surgery; [4] no cardiac activity was present in the EP if US was performed before the procedure; [5] there was no evidence in the medical record of hepatic dysfunction or renal disease; [6] complete blood count did not suggest a blood dyscrasia; and [7] there was no evidence in the medical record of a lack of reliability or poor compliance (that is, intra- venous drug use, failure to keep appointments before the diagnosis, short-term incarceration). Hospital charges associated with each case were computed from billing statements. Surgical professional fees were $1,647 in 1991 at our institution for EP regardless ofthe procedure(s} performed. Anesthesia professional fees were estimated from the surgical procedure, billed anesthesia time, and American Society of Anesthesiologists physical status. The hospital charges and professional fees were used to estimate direct hospital costs; indirect costs (lost productivity or the value of output foregone by women suffering from EP) were estimated according to published data indicating the likelihood of employment and average salary based on age (7). Disability was factored at 28 days for laparotomy, 7 days for laparoscopy, and 4 days for MTX treatment. A cost analysis was completed comparing two management policies: [1] surgical management in all patients with "standard treatment" for our institution: exam under anesthesia, culdocentesis, dilation and curettage (D and C), and/or laparoscopy as needed to confirm the diagnosis of EP, then salpingostomy or salpingectomy via laparotomy and [2] medical management for patients eligible for MTX treatment and appropriate surgical treatment for the others. Medical management was based on the protocol for single-dose MTX as reported by Stovall and co-workers (6) (excluding the i3-hcg on day 2). In Stovall's report, 95% of cases had a i3-hcg of zero within 50 days (6); thus, i3-hcg follow-up in this analysis is included through day 49. Costs for MTX failures include the cost ofmtx treatment, the average total direct cost for laparotomy in patients not eligible for MTX, and a total of 28 days of disability. RESULTS Fifty-three cases of EP were identified; upon review ofthe medical records, 50 patients with 51 EPs were confirmed. Among the 51 cases, 50 received standard treatment for our institution; diagnostic laparoscopy was performed in 22 (44%) of these patients. One patient underwent diagnostic laparoscopy with a finding of a 2.5-cm right isthmic EP followed by MTX 50 mg/m 2 1M (as described by Stovall and co-workers) (6). This patient, whose MTX treatment was successful, is not included in the cost analyses because she did not receive standard therapy. 964 Creinin and Washington Cost of ectopic treatment Fertility and Sterility

3 Table 1 Age, (y) 15 to to to to to to 44 Average Parity o 1 Patient Characteristics ~2 Risk factors* EP Clinical PID High-risk sexual behavior without a history of clinical PID IUD Tubal ligation Infertility None reported Ectopic location Tubal Tubal abortion Ruptured Cornual Ruptured * Values in parentheses are percents. t Patients may have more than one risk factor. No. of patients 5 (10)* 10 (20) 21 (42) 6 (12) 6 (12) 2 (4) 27.9 ± (36) 17 (34) 15 (30) 8 (16) 8 (16) 5 (10) 3 (6) 2 (4) 2 (4) 28 (56) Patient characteristics are described in Table 1. Three of the patients with tubal pregnancies were hemodynamically unstable on presentation to the emergency room. More than 25% of these patients either had a history of clinical pelvic inflammatory disease (PID) or were known to have practices consistent with high-risk sexual behavior (intravenous drug abuse, prostitution). Additionally, >50% had last menstrual periods that were either unknown or unsure. Fifteen (30%) patients were eligible for MTX therapy. Hospital stay totaled 3 days for 8 cases and 4 days for 24 cases; the remaining 18 (36%) were hospitalized for >4 days. Twenty percent of patients had a temperature during the postoperative period >38.5 C, 20% were treated with intravenous antibiotics, 14% had a urinary tract infection, 12% required a transfusion, 4 % had urinary retention or incontinence, 4% had a wound infection, 2% had a postoperative ileus, and 2% had persistent trophoblast requiring reoperation. Potentially, with MTX therapy in appropriate patients, 22% of prolonged hospital stays could have been avoided, as could have 20% of the patients with fever >38.5 C, 20% of the patients requiring intravenous antibiotics postoperatively, and the one patient who had a second operative procedure and hospitalization for persistent trophoblast. Hospital (direct) charges are summarized in Table 2. Both operating room and anesthesia costs and anesthesia professional fees are less for patients who were not MTX eligible as compared with those who were eligible. Because the patients who were not MTX eligible included the patients with larger (more easily palpable on exam) and ruptured EPs, these patients were less likely to undergo D and C (23% versus 53%) or diagnostic laparoscopy (31% versus 67%) in the operating room, thereby decreasing time in the operating room and surgical and anesthetic costs. The approximate cost of MTX treatment at our institution in 1992 was $1,495. This total is based on the protocol for single-dose MTX as reported by Stovall and co-workers (6) including an Emergency Department visit (average $315), an endovaginal US, D and C, before and after-treatment blood tests, MTX, two follow-up office visits, and ~-hcg follow-up. Because patients with a ~-hcg >2,000 miu /ml with no intrauterine pregnancy on endovaginal US do not need a D and C (6), this would further reduce the cost in these cases to $1,369. The total direct cost of EP at our institution was $525,464. The costs of diagnosis (for example, serum ~-hcg, US when needed) are included in the direct costs if they were done in the Emergency Department when the patient presented for treatment; diagnostic tests and procedures performed before that time are not included. Fifty-two percent of our patients had not been evaluated at our institution for the EP before diagnosis. Lost wages were estimated as $93,553 and lost value of household management estimated as $43,163 for a total indirect cost of $136,716; thus, the total annual cost for EP was $662,180. With MTX treatment of the 30% who were deemed retrospectively to be eligible, the direct costs would total $387,160 and indirect costs $101,560 for a total of $488,720. Factoring in one MTX treatment failure (a 93% success rate), the direct costs would total $397,561 and indirect costs $103,904 for a total of $501,465; this is a cost savings of 24 % compared with surgical management. Assuming that 30% of all EPs could be treated with single-dose systemic MTX, the total cost savings with MTX therapy can be extrapolated to a Creinin and Washington Cost of ectopic treatment 965

4 u! kltlh Table 2 Direct Costs* MTX eligible MTX NotMTX Cases without Cases with with All cases eligible eligible complications complications complications (n = 50) (n = 15) (n = 35) (n = 26) (n = 24) (n = 7) Hospital costs Operating and recovery room 2,015 ± 475 2,216 ± 600 1,928 ± 390 2,032 ± 485 1,996 ± 474 2,106 ± 588 Anesthesia 1,149 ± 407 1,388 ± 410 1,046 ± 365 1,143 ± 427 1,156 ± 393 1,243 ± 388 Room 2,661 ± 2,024 2,307 ± 1,171 2,813 ± 2,294 2,019 ± 681 3,357 ± 2,694 2,746 ± 1, Laboratory 918 ± ± ± ± 384 1,099 ± ± Emergency departmentt 315 ± ± ± ± ± ± 175 Pharmacy 404 ± ± ± ± ± ± 346 Total hospital charges:j: 8,045 ± 2,936 8,179 ± 2,299 7,988 ± 3,199 6,860 ± 1,143 9,330 ± 3,692 9,277 ± 2,765 (4,629 to 21,148) (4,629 to 14,502) (5,144 to 21,148) (4,629 to 9,104) (5,144 to 21,148) (6,046 to 14,502) Professional costs Surgon's fee Anesthesiologist's fee 817 ± ± ± ± ± ± 147 Total inpatient charges 10,509 ± 2,943 10,716 ± 2,283 10,421 ± 3,210 9,318 ± 1,237 11,800 ± 3,668 11,774 ± 2,704 (8,543 to 16,849) (7,226 to 23,445) (7,226 to 16,849) (7,458 to 23,445) (7,226 to 12,101) (7,491 to 23,445) * Values are $ average ± SD. t Not all cases had emergency department charges; the figures only include those cases that incurred an emergency department charge. :j: Also included in the total hospital charges are charges forradiology, blood products, general supplies, and posthospitalization follow-up (for example, serum f3-hcg, dressings and supplies for open wounds, re-hospitalization). Values in parentheses are ranges. national level. The potential annual cost savings could approach >$281.9 million with a 95% success rate and $265.0 million with a 90% success rate. Because different patient populations will have varying proportions of patients eligible for MTX therapy, the variation in national cost savings are estimated in Figure 1. DISCUSSION This is the first analysis of the potential cost savings with MTX as a management option for EP. This report has described the tremendous potential for cost reduction at a county hospital with an alternative nonsurgical treatment for selected cases of EP. We found that 30% of our population, in retrospect, would have been eligible for treatment with MTX; because the eligibility criteria are evaluated retrospectively, some patients may have been included or excluded inappropriately. The eligibility rate in our population is almost identical to the 33.4% (109/326) that Stovall et al. (10) found were eligible over a prospective period of 28 months ending in February However, Stovall and Ling (11) found 44.8% of patients eligible over the next 26 months ending in April 1992, 14 (5.2%) of which had cardiac activity on pretreatment US. Because this report is retrospective, it cannot measure the true cost savings but does approximate the potential. One cost of MTX therapy that cannot be estimated because of this report's retrospective analyses is the cost of hospitalization for observation and US examinations when pain is experienced during resorption of the trophoblast (6); because this occurs in a minority of patients, this additional cost would only minimally detract from the cost savings. A prospective analysis would be necessary to accurately determine the financial benefits of MTX treatment in appropriately selected patients in terms of both hospital costs and disability /lost wages. Because the charges at a private hospital may be higher than those at a public/county facility, the savings from MTX treatment may actually be greater. In the private setting, the patient may be more likely to be diagnosed in the office rather than requiring an emergency room visit; this would also increase the cost savings. Approximately half of the patients treated surgically had some costs of diagnosis not included in the cost analysis; because the cost of diagnosis is included in the analysis for 966 Creinin and Washington Cost of ectopic treatment Fertility and Sterility

5 III a:..j '"..J 0 Q Z 0 :::;..J :Ii O~~~-T~~~~~~~-T~~~ o % ELIGIBLE FOR METHOTREXATE Figure 1 Annual national savings with MTX treatment of EP in appropriate patients as compared with surgical treatment. D, total cost (95% success); 0, total cost (90% success); 0, direct costs (95% success); ~, direct costs (90% success); 0, indirect costs (95% success);., indirect costs (90% success). MTX therapy, the actual cost savings for MTX therapy may be even greater. In lower risk populations in whom MTX therapy may be an appropriate option for a greater percentage of cases, savings will be greater. Although it is not the focus of our study, laparoscopy, when used not only for diagnostic purposes but also for treatment of EP, would result in further savings. Operative laparoscopy is well documented to be efficacious in the treatment of EP (3, 4, 8, 12-17). For an estimate of cost savings within our population, one can assume that laparoscopy could have been used for treatment in the 25 patients not eligible for MTX and with an unruptured tubal pregnancy. For a direct cost savings of $1,500 per case as compared with laparotomy (4) and a 93% success rate for MTX treatment, the direct costs would total $360,061 and indirect costs $52,633 for a total of $412,694; this is a cost savings of 38% compared with management solely by laparotomy. With more experience, the gynecological surgeon would be able to treat some ruptured tubal pregnancies laparoscopically, further decreasing morbidity and cost. The main concern with interpretation of these results is the assumption that MTX has definitively proven effectiveness in the treatment of selected cases of EP; this concern, notwithstanding, the data are encouraging. Overall, 13 series of investigational studies have reported on >250 selected cases of tubal pregnancy safely and effectively treated with systemic MTX. The peak serum level of MTX after intramuscular injection for EP is significantly lower than the accepted toxic level (18). Additionally, patients treated for EP with MTX frequently can continue to work during therapy or return to work in less than 1 week; in contrast, return to work usually takes 4 to 6 weeks after laparotomy and 1 to 2 weeks after operative laparoscopy. Follow-up studies have demonstrated that MTX does not impair future fertility; menses return normally and pregnancy rates are at least as good as those achieved by traditional surgical treatment of EP (10, 19). However, there are no studies that address future fertility in patients treated surgically with EPs <3.5 cm only, the maximum size for which women would be eligible for treatment with MTX. In May 1991, Stovall and co-workers (6) reported the successful use of MTX for EP in a single dose 1M treatment without leucovorin rescue. This regimen has the benefits of single-dose treatment, high efficacy, low morbidity, and low cost. Stovall and Ling (11) recently reported the use of the singledose regimen in 121 patients; a 94.3% success rate was achieved, which is equal in efficacy to multidose regimens involving leucovorin rescue. We routinely have been using this treatment for EP in appropriately selected patients since Attempts to use MTX locally as an intratubal injection have had mixed results. The prospective success rate appears to be approximately 80% (20); however, a recent randomized prospective trial of intratubal MTX versus laparoscopic salpingostomy was abandoned after only 12 patients because of poor results with the intratubal MTX (21). Because the efficacy rate in appropriately selected patients of systemic MTX therapy (94% to 95%) is greater than that of intratubal injection, the less invasive intramuscular route is the more appropriate choice. The efficacy of MTX therapy for EP is comparable with conservative surgical therapy (salpingostomy). Persistent trophoblast remains after laparoscopic salpingostomy in 0% to 20.0% of cases reported, with a cumulative persistence rate of6.8% Creinin and Washington Cost of ectopic treatment 967

6 (4,8, 12-17,22-24). With salpingostomy via laparotomy, persistent trophoblast remains in 0% to 6.5% of cases reported with a cumulative rate of 2.0% (4, 8, 16, 17, 23-25). The persistence rate in our series was 8.5% (12 salpingostomies via laparotomy). Interestingly, MTX has been used for treatment of persistent trophoblast after salpingostomy. Since 1986, 12 reports have described 36 cases treated with oral, intramuscular, or intravenous MTX in various regimens; 33 of 35 (94.3%) were successful. Thus, approximately 5% to 6% of patients treated with MTX or conservative surgical management fail therapy and are at risk for tubal rupture and hemorrhage; both methods of treatment require close patient follow-up. Likewise, it is important to closely monitor those patients treated with MTX for persistent trophoblast; the success rate in limited reports is approximately the same as when MTX is used for primary therapy. Thus, MTX may provide an alternative to surgery for treatment of appropriately selected patients with EP. This medical management in a single-dose regimen of 50 mg/m 2 appears to be as equally effective as conservative laparoscopic management but with less morbidity and cost. Ideally, patients could be diagnosed with an EP based on history, physical examination, laboratory, and US evaluations. For those patients who meet the criteria for treatment with MTX, such therapy can be used. Patients who do not meet the criteria or in whom the diagnosis is unsure should be treated laparoscopically. Because MTX is not necessarily a better treatment than laparoscopic salpingostomy or salpingectomy, a patient requiring laparoscopy for diagnosis should be surgically treated via the laparoscope. At this time, it makes no sense to diagnose the EP with laparoscopy and then treat the patient with MTX (if the tubal mass is ::;;3.5 cm in diameter) unless the operator is unskilled at operative laparoscopy. In that case, treatment with MTX could be considered less morbid and more cost-effective than laparotomy. Importantly, key outcomes such as the rate of ectopic rupture and subsequent death remain to be determined. Consequently, the use of MTX for EP should still be considered investigational; we recognize and agree that the evolution of this treatment option should be based primarily on its ability to decrease morbidity and mortality rather than on its ability to reduce costs. If proven to be effective, the potential cost savings demonstrated in this report will become an important factor in selecting among alternative regimens. Acknowledgment. We thank Michelle Berlin, M.D., for her editorial assistance. REFERENCES 1. Centers for Disease Control. Ectopic pregnancy-united States, MMWR 1992;41: Koonin LM, Atrash HK, Lawson HW, Smith JC. Maternal mortality surveillance, United States, MMWR CDC Surveill Summ 1991;40: Murphy AA, Nager CW, Wujek JJ, Kettel LM, Torp VA, Chin HG. Operative laparoscopy versus laparotomy for the management of ectopic pregnancy: a prospective trial. Fertil Steril 1992;57: Vermesh M, Silva PD, Rosen GF, Stein AL, Fossum GT, Sauer MV. Management of unruptured ectopic gestation by linear salpingostomy: a prospective, randomized clinical trial of laparoscopy versus laparotomy. Obstet Gynecol 1989;73: Maymon R, Shulman A, Maymon BB, Bar-Levy F, Lotan M, Bahary C. Ectopic pregnancy, the new gynecological epidemic disease: review of the modern work-up and the nonsurgical treatment options. Int J Fertil 1992;37: Stovall TG, Ling FW, Gray LA. Single-dose methotrexate for treatment of ectopic pregnancy. Obstet Gynecol 1991;77: Washington AE, Katz P. Ectopic pregnancy in the United States: economic consequences and payment source trends. Obstet Gynecol 1993;81: Baumann R, Magos AL, Turnbull A. Prospective comparison of videopelviscopy with laparotomy for ectopic pregnancy. Br J Obstet GynaecoI1991;98: Division of Quality Control Management, American Hospital Association. International Classification of Diseases, Ninth Revision, Clinical Modification. Chicago (ILj: American Hospital Publishing Inc., Stovall TG, Ling FW, Gray LA, Carson SA, Buster JE. Methotrexate treatment of un ruptured ectopic pregnancy: a report of 100 cases. Obstet GynecoI1991;77: Stovall TG, Ling FW. Single dose methotrexate: an expanded clinical trial. Am J Obstet Gynecol 1993;168: Pouly JL, Mahnes H, Mage G, Canis M, Bruhat MA. Conservative laparoscopic treatment of 321 ectopic pregnancies. Fertil Steril1986;46: DeCherney AH, Diamond MP. Laparoscopic salpingostomy for ectopic pregnancy. Obstet Gynecol 1987;70: Silva PD. A laparoscopic approach can be applied to most cases of ectopic pregnancy. Obstet GynecoI1988;72: Henderson SR. Ectopic tubal pregnancy treated by operative laparoscopy. Am J Obstet Gynecol 1989;160: Brumsted J, Kessler C, Gibson C, Nakahima S, Riddick DH, Gibson M. A comparison of laparoscopy and laparotomy for the treatment of ectopic pregnancy. Obstet Gynecol 1988;7: Zouves C, Urman B, Gomel V. Laparoscopic surgical treat- 968 Creinin and Washington Cost of ectopic treatment Fertility and Sterility

7 ment of tubal pregnancy. A safe, effective alternative to laparotomy. J Reprod Med 1992;37: Schiff E, Shalev E, Bustan M, Tsabari A, Mashiach S, Weiner E. Pharmacokinetics of methotrexate after local tubal injection for conservative treatment of ectopic pregnancy. Fertil Steril 1992;57: Stovall TG, Ling FW, Buster JE. Reproductive performance after methotrexate treatment of ectopic pregnancy. Am J Obstet Gynecol 1990;162: Pansky M, Bukovsky I, Golan A, Herman A, Hertziano I, Langer R, et al. Methotrexate local injection for unruptured tubal pregnancy: an alternative to laparotomy? Int J Gynecol Obstet 1992;37: Mottla GL, Rulin MC, Guzick DS. Lack of resolution of ectopic pregnancy by intratubal injection of methotrexate. Fertil Steril1992;57: Vermesh M, Silva PD, Sauer MV, Vargyas JM, Lobo RA. Persistent tubal ectopic gestation: patterns of circulating J3-human chorionic gonadotropin and progesterone, and management options. Fertil Steril 1988;50: Lundorff P, Hahlin M, Sjoblom P, Lindblom B. Persistent trophoblast after conservative treatment of tubal pregnancy: prediction and detection. Obstet Gynecol 1991;77: Seifer DB, Gutmann IN, Grant WD, Kamps CA, DeCherney AH. Comparison of persistent ectopic pregnancy after laparoscopic salpingostomy versus salpingostomy at laparotomy for ectopic pregnancy. Obstet GynecoI1993;81: DiMarchi JM, Kosasa TS, Kobara TY, Hale RW. Persistent ectopic pregnancy. Obstet Gynecol 1987;70: Note. References for the use of MTX for persistent trophoblast are available on request from the authors. Creinin and Washington Cost of ectopic treatment 969