HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hardesty, L. A. Articles by Kelley, J. L. Search for Related Content PubMed PubMed Citation Articles by Hardesty, L. A. Articles by Kelley, J. L.

Use of Preoperative MR Imaging in the Management of Endometrial Carcinoma: Cost Analysis¹

¹ From the Department of Radiology, Magee Women's Hospital, University of Pittsburgh, 300 Halket St, Pittsburgh, PA 15213. From the 1998 RSNA scientific assembly. Received December 2, 1998; revision requested February 15, 1999; final revision received July 1; accepted July 26. Address reprint requests to L.A.H. (e-mail: lhardesty@mail.magee.edu).

	Abstract

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
PURPOSE: To compare the cost of magnetic resonance (MR) imaging and its ability to direct the use of lymph node dissection with the cost and ability of conventional surgery for the staging of endometrial carcinoma.

MATERIALS AND METHODS: Preoperative MR images of 25 patients who underwent hysterectomy for endometrial carcinoma were retrospectively evaluated. MR imaging results were compared with those of intraoperative gross dissection of the uterus and final histopathologic examination. Medicare reimbursements for two scenarios were compared in each patient. In the MR imaging scenario, the necessity for lymph node dissection was based on MR imaging results and histologic findings at biopsy. In the actual scenario, lymph node dissection was performed at the surgeon's discretion on the basis of findings at gross dissection of the uterus and histologic examination at biopsy.

RESULTS: The cost of the MR imaging scenario, as defined by Medicare reimbursements, was 1% ($1,265/$148,500) less than that of the actual scenario. In the MR imaging scenario, all patients who required lymph node dissection received it, and 86% of the lymph node dissections performed were necessary. In the actual scenario, one necessary lymph node dissection was not performed, and only 31% of the lymph node dissections performed were necessary.

CONCLUSION: Staging with MR imaging has costs and accuracy similar to those of the current method of staging with intraoperative gross dissection of the uterus. In addition, MR imaging decreases the number of unnecessary lymph node dissections.

Index terms: Cost-effectiveness, 854.32 • Lymphatic system, neoplasms, 99.8312 • Magnetic resonance (MR), comparative studies • Magnetic resonance (MR), contrast enhancement, 854.121411, 854.121412, 854.12143 • Uterine neoplasms, metastasis, 99.8312 • Uterine neoplasms, MR, 854.121411, 854.121412, 854.12143, 854.32 • Uterine neoplasms, staging, 854.32 • Uterus, endometrium, 854.32

	Introduction

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
Endometrial carcinoma is the most common gynecologic malignancy and the fourth most common carcinoma in female patients (1). In the United States in 1998, 36,100 new cases and 6,300 deaths due to endometrial carcinoma occurred (2). The number of patients will continue to rise as life expectancy is prolonged because 75% of these patients are older than 50 years (1).

Vaginal bleeding in postmenopausal women is the most common presentation of endometrial carcinoma. Findings at endocervical curettage and suction endometrial biopsy confirm the diagnosis of endometrial carcinoma and provide the tumor grade and histologic type. Surgical staging with the International Federation of Gynecology and Obstetrics (FIGO) system (3) (Table 1) is performed because staging at clinical examination results in the understaging of up to 22% of cases (4). At surgical staging, a total abdominal hysterectomy is performed, and the uterus is sectioned in the operating room. The surgeon awaits the results of the pathologist's evaluation of the specimen (Fig 1). Pelvic and paraaortic lymph nodes are sampled at the discretion of the surgeon after he or she considers the initial histopathologic findings.

View larger version (113K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Intraoperatively obtained gross specimen of the uterus, bivalved in a sagittal plane, shows deep invasion (>50% of the myometrial thickness) of endometrial carcinoma (arrows).

Preoperative imaging in patients with endometrial carcinoma varies by institution. Although magnetic resonance (MR) imaging has been shown to be accurate in the assessment of myometrial invasion, its use has not been universally accepted (7–10). Previous studies of the ability of MR imaging to depict the depth of myometrial invasion were performed by comparing the results at MR imaging with the final results at histopathologic examination. The accuracy of MR imaging in the evaluation of the depth of invasion is 75%–87% (7–10). But final histopathologic findings are not available when the surgeon must decide whether lymph node sampling is necessary.

In most institutions, the decision to perform lymph node sampling is currently made on the basis of the preoperative histologic results and the depth of tumor invasion as determined at intraoperative gross dissection of the uterus. If preoperative MR imaging is shown to be as accurate as intraoperative gross dissection, intraoperative uterine dissections and unnecessary lymph node dissections could be eliminated.

We undertook this retrospective study to compare the accuracy of preoperative MR imaging in the direction of the appropriate use of lymph node dissection with that of conventional intraoperative gross evaluation of the uterus. We also compared the cost (defined as Medicare reimbursement) of this application of MR imaging with that of conventional surgical staging.

	MATERIALS AND METHODS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
All 34 patients with primary untreated endometrial carcinoma who underwent preoperative MR imaging at our institution from January 1995 to March 1998 were considered for this study. The histologic diagnosis of endometrial carcinoma was made on the basis of findings at endometrial biopsy. Six patients were treated with radiation therapy or chemotherapy prior to or in place of surgery and were excluded from the study. Three patients were found to have mixed müllerian tumors at final histopathologic examination and were excluded from the study.

The remaining 25 patients (age range, 35–83 years; mean age, 59 years) formed the basis of this study. Of the 25 patients, 22 (88%) had endometrioid adenocarcinoma, two (8%) had serous carcinoma, and one (4%) had clear cell carcinoma. The FIGO stages were as follows: one (4%) stage Ia, 16 (64%) stage Ib, five (20%) stage Ic, two (8%) stage IIa, and one (4%) stage IIb. There were no FIGO stage III or IV carcinomas because patients with these received chemotherapy or radiation therapy either prior to or in place of hysterectomy and were disqualified from this study. Despite the retrospective nature of the study, the histopathologic findings and stages in the patients examined were representative of the epidemiologic data for endometrial carcinoma (1,11).

MR imaging was performed with a 1.5-T superconducting magnet (Horizon; GE Medical Systems, Milwaukee, Wis). Data were collected with a 256 x 192 matrix and were displayed in a 256 x 256 matrix. By using the body coil, T1-weighted spin-echo (400–800/10 [repetition time msec/echo time msec]) images with a 10-mm section thickness and a 2-mm intersection gap were obtained in coronal plane from the diaphragm to the pubic symphysis and in the transverse plane from the diaphragm to the iliac crests.

Imaging with the remaining sequences was performed with a pelvic phased-array coil and 5-mm section thickness with a 2-mm intersection gap. Transverse T1-weighted spin-echo (400–800/10) pelvic images and transverse and sagittal T2-weighted fast spin-echo (3,000–3,950/76) pelvic images and sagittal T2-weighted fast spin-echo (2,000–6,000/75–100; echo train length of eight) pelvic images were obtained.

Then, 0.1 mmol of gadopentetate dimeglumine (Omniscan; Nycomed, Princeton, NJ) per kilogram of body weight was injected intravenously, and sagittal dynamic T1-weighted spoiled gradient-recalled acquisition in the steady state, or GRASS (GE Medical Systems), (9.4/4.2, flip angle of 70°) images of the uterus were obtained every 45 seconds for 3 minutes. Transverse contrast material–enhanced fat-saturated T1-weighted spin-echo (400–800/10) images of the pelvis were then obtained.

The first five patients in this study did not receive intravenous injections of contrast material during MR imaging because in 1995 and early 1996 the administration of gadolinium-based contrast material was not a component of the standard protocol at our institution for MR imaging of endometrial carcinoma.

The average time between MR imaging and hysterectomy was 7.6 days (range, 1–34 days). The MR images were reviewed independently and retrospectively by three radiologists (J.H.S., T.S.C., C.M.J.) who were experienced in the interpretation of pelvic MR images. Readers were blinded to all information except that the patients had endometrial carcinoma that was proved at biopsy. Readers were asked to evaluate the depth of myometrial invasion as superficial (<50% of myometrial thickness, FIGO stages Ia and Ib) or deep (50% of myometrial thickness, FIGO stage Ic) and to note the presence of tumor within the cervical parenchyma. The interpretations of the three readers were tabulated, and a single depth, determined by the majority opinion, was reported for each patient.

Patients underwent hysterectomy and intraoperative gross dissection of the uterus. Lymph node sampling was performed at the discretion of the surgeon. The surgeon's decision was directed by the preoperative histologic results obtained from suction endometrial biopsy and by the results of intraoperative gross dissection of the uterus. The depth of myometrial invasion and the presence of cervical involvement determined by the pathologist at intraoperative gross dissection were retrospectively obtained from the written pathology report provided at the time of surgery. The interpretations at MR imaging and gross dissection were compared with the interpretation in the final written pathology report.

The accuracy in the indication of lymph node dissection and the costs associated with two practice scenarios were determined in all patients. In the MR imaging scenario, lymph node dissection was performed if serous, clear cell, or grade 3 endometrioid histologic findings were present or if MR imaging revealed deep myometrial invasion or cervical invasion. In the actual scenario, lymph node dissection was performed if serous, clear cell, or grade 3 endometrioid histologic findings were present of if intraoperative gross dissection of the uterus showed deep myometrial invasion or cervical invasion. Several surgeons performed lymph node dissections at their discretion despite findings of superficial invasion of the myometrium by tumor at gross dissection of the uterus.

The cost of each procedure was approximated by the Medicare reimbursement rate (12). The amount of Medicare reimbursement for each procedure was determined by use of the Current Procedural Terminology (CPT) and diagnosis-related groups (DRG) codes (Table 2). For outpatient procedures, reimbursement included a professional fee and a technical component fee for each CPT code. CPT codes were also used to determine the reimbursement rate for the professional fee for each inpatient procedure. There was no technical component for the inpatient procedures, as that reimbursement is determined by use of the DRG codes.

	RESULTS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
With preoperative MR imaging, the depth of myometrial invasion was correctly categorized in 22 (88%) patients and overestimated in three (12%) patients. Superficial invasion (84% specificity) was correctly categorized in 16 of 19 patients, and deep invasion was correctly categorized in six of six patients (100% sensitivity) (Fig 2a, Table 3). Superficial invasion was incorrectly characterized as being deep in three patients, but the tumor in one of these patients was clear-cell type, so the patient was at high risk for lymph node metastases, and lymph node dissection was required regardless of the interpretation at MR imaging. One patient had tumor extension into the cervical parenchyma (FIGO stage IIb) that was correctly diagnosed at MR imaging (Fig 2b). Two patients had microscopic cervical extension (FIGO stage IIa) that was not diagnosed at MR imaging.

View larger version (165K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. (a) Transverse T2-weighted fast spin-echo (3,950/76) image shows deep invasion (>50% of the myometrial thickness) of endometrial carcinoma (arrow). (b) Transverse T2-weighted fast spin-echo (3,000/76) image obtained in the same patient shows that the endometrial carcinoma (arrow) involves the cervix.

View larger version (162K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. (a) Transverse T2-weighted fast spin-echo (3,950/76) image shows deep invasion (>50% of the myometrial thickness) of endometrial carcinoma (arrow). (b) Transverse T2-weighted fast spin-echo (3,000/76) image obtained in the same patient shows that the endometrial carcinoma (arrow) involves the cervix.

The patient in whom the depth of invasion was overestimated at gross dissection was not one of the patients in whom invasion was overestimated at MR imaging. In patients in whom the depth of invasion was correctly determined at MR imaging, the depth was incorrectly determined at gross dissection, and vice versa. The single patient with parenchymal cervical involvement was correctly identified at both MR imaging and gross dissection. Both MR imaging and gross dissection failed in the identification of two patients with microscopic cervical involvement, but both patients underwent the appropriate lymph node dissection for other reasons. In one patient, a grade 3 endometrioid histologic type indicated dissection, while in the other patient, deep myometrial invasion that required lymph node dissection was correctly identified at both MR imaging and gross dissection.

Staging at MR imaging and initial histologic evaluation would have indicated lymph node dissection in 13 of the 25 patients (four on the basis of histologic findings; eight, deep myometrial invasion; and one, parenchymal cervical invasion). Lymph node dissection would have been appropriate in 11 of the 25 patients (Table 4). In two patients, carcinoma was overstaged at MR imaging, and the patients would have undergone unnecessary lymph node dissections. Tumor in the third patient in whom carcinoma was overstaged at MR imaging had a serous histologic type and would have required lymph node dissection regardless of the depth of invasion.

Alternatively, 20 of these 25 patients actually underwent lymph node dissection, which was indicated by the initial histologic findings, by findings at intraoperative gross dissection of the uterus, or by the surgeon's discretion. The lymph node dissections were appropriate in 11 of the 25 patients (four on the basis of histologic findings; six, deep myometrial invasion; and one, parenchymal cervical invasion) (Table 4). Nine of the 25 patients with a favorable cell type at histologic findings and findings of superficial tumor at gross dissection underwent unnecessary lymph node dissections at the discretion of the surgeon.

The total Medicare reimbursement for the actual scenario was $148,500. This was calculated as follows: Twenty-five intraoperative gross dissections x $147 = $3,675; 25 hysterectomies x $1,137 = $28,425; 20 lymph node dissections x $1,320 = $26,400; and 25 inpatient admissions for endometrial carcinoma x $3,600 = $90,000.

The Medicare reimbursement would have been $1,265 more for the actual scenario than for the MR imaging scenario. This represented a difference of less than 1% ($1,265/$148,500), or $51, per patient.

	DISCUSSION

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
In an era of increasing cost-consciousness, an imaging test is used if it provides, at a reasonable cost, information that can change patient treatment. Although MR imaging has been shown to be relatively accurate in the assessment of the depth of myometrial invasion (6,13–16), gynecologists have not embraced MR imaging for preoperative staging of endometrial carcinoma. Most gynecologic surgeons believe that, although their initial clinical examination is somewhat limited, gross dissection of the uterus gives the same, if not more reliable, information than does MR imaging at a substantially lower cost. The comparison of Medicare reimbursements for the two treatment scenarios in this study showed that their costs were similar; the cost of the preoperative MR imaging scenario was 1% less than the cost of the conventional surgical staging scenario.

The findings from this study show that the accuracy of preoperative MR imaging is essentially equivalent to that of intraoperative gross dissection of the uterus in the determination of the depth of tumor invasion. It has been reported that, when the findings are incorrect, the depth of myometrial invasion is usually overestimated at MR imaging (13), as it occurred in three patients in this study. When MR imaging findings are incorrect, they usually occur in patients with adenomyosis, an intramural or subendometrial myoma, a polypoid tumor, a thin atrophic myometrium, or an atrophic junctional zone between the endometrium and the myometrium (7,13,14,17).

Our study showed similar findings. One patient in whom the depth of invasion was overestimated at MR imaging had extensive adenomyosis that all three radiologists commented on in their interpretations. Final histopathologic examination revealed only microscopic tumor invasion in this patient. Another patient in whom the depth of invasion was overestimated at MR imaging had subserosal and intramural leiomyomas up to 4 cm in diameter; final histopathologic examination showed 40% myometrial invasion. In the third patient in whom the depth of invasion was overestimated at MR imaging, histopathologic examination revealed that the tumor was composed of friable papillary excrescences that invaded 10% of the myometrial thickness.

Although the depth of myometrial invasion was overestimated at MR imaging in three patients, tumor extension was not underestimated at MR imaging in any patient. Therefore, no patient at high risk for lymph node metastases would have been inappropriately denied lymph node dissection. In contrast, at intraoperative gross dissection of the uterus, one case of deep invasion was incorrectly categorized as being superficial. This patient's final histopathology report revealed that the tumor penetrated deeper (60%) into the myometrium at microscopic evaluation than at gross evaluation (<50%). Inaccurate interpretation can occur at gross dissection when the bulk of the tumor is limited to the inner half of the myometrium, with only microscopic islands of deeply invasive tumor (4). In general, we expect MR imaging to have this same limitation, although it did not in this particular case. As with MR imaging, adenomyosis and leiomyomas that distort the uterine wall are other causes of misinterpretation of tumoral depth at intraoperative gross dissection (4).

Given the similar accuracy and Medicare reimbursement rate for the MR imaging scenario and the traditional surgical staging scenario, we postulate that preoperative tumor staging at MR imaging has potential advantages. For example, patients would be better informed about the operation they would undergo. Operating room scheduling would be more efficient, with time for lymph node dissection reserved for only those patients who actually require it rather than for all patients with endometrial carcinoma, as it is now done. Patients would be referred to a physician with the appropriate level of surgical expertise. If lymph node dissection were not required, the patient's gynecologist could perform the hysterectomy. Last, and perhaps most important, the performance of fewer unnecessary node dissections would decrease the morbidity associated with lymphadenectomy (5).

Surgeons, compelled to minimize operating time, have an incentive to proceed with lymph node dissection while they await the findings at gross dissection of the uterus. In this study of 25 patients, this practice led to the performance of nine unnecessary lymph node dissections at the discretion of the surgeon, despite a favorable histologic type at biopsy and findings of superficial invasion at gross dissection of the uterus. In contrast, MR imaging findings led to the performance of only two unnecessary lymph node dissections.

This study has weaknesses, which include its retrospective nature and small size. In retrospect, it is unclear how the gynecologic oncologists selected patients for preoperative MR imaging. All of the 34 patients with endometrial carcinoma who underwent MR imaging in the 39 months of this study were initially included. Another weakness is that the surgeons were not blinded to the MR imaging results; this may have affected their decisions to perform lymph node dissection when the findings at gross dissection indicated that it was not necessary. Five of the early MR imaging examinations were performed without contrast enhancement, which may have influenced the radiologists' interpretations.

MR imaging can be used to preoperatively stage endometrial cancer; its cost, as defined by Medicare reimbursement, and ability are similar to those of conventional surgical staging with intraoperative uterine dissection. The additional advantages of preoperative MR imaging, especially the decreased number of unnecessary lymph node dissections, make it a valuable tool in the preoperative assessment of patients with endometrial carcinoma. The findings from this small study suggest that preoperative MR imaging may eventually replace intraoperative gross dissection of the uterus in the selection of patients with endometrial carcinoma for lymph node dissection.

	Acknowledgments

 
The authors gratefully acknowledge the editorial assistance of Judith R. Lave, PhD, Professor of Health Economics at the University of Pittsburgh Graduate School of Public Health, Pa.

	Footnotes

 
Abbreviations: CPT = Current Procedural Terminology DRG = diagnosis-related groups FIGO = International Federation of Gynecology and Obstetrics

Author contributions: Guarantors of integrity of entire study, L.A.H., J.H.S.; study concepts and design, J.H.S., L.A.H.; definition of intellectual content, L.A.H., J.H.S.; literature research, L.A.H., F.V.P., J.L.K.; clinical studies, all authors; data acquisition, L.A.H.; data analysis, L.A.H., J.H.S.; manuscript preparation, editing, and review, L.A.H., J.H.S.

	References

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Barakat RR, Park RC, Grigsby PW, Muss HD, Norris HJ. Corpus: epithelial tumors. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia, Pa: Lippincott-Raven, 1997; 859-896.
2. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 1998. CA Cancer J Clin 1998; 48:6-29.[Abstract]
3. Creasman W, Odicino F, Maisonneuve P, et al. Carcinoma of the corpus uteri. J Epidem Biostat 1998; 1:35-61.
4. Hricak H, Rubinstein LV, Gherman GM, Karastaedt N. MR imaging evaluation of endometrial carcinoma: results of an NCI cooperative study. Radiology 1991; 179:829-832.[Abstract/Free Full Text]
5. Larson DM, Johnson K, Olson KA. Pelvic and para-aortic lymphadenectomy for surgical staging of endometrial cancer: morbidity and mortality. Obstet Gynecol 1992; 79:998-1001.[Medline]
6. Doering DL, Barnhill DR, Weiser EB, Burke TW, Woodward JE, Park RC. Intraoperative evaluation of depth of myometrial invasion in stage I endometrial adenocarcinoma. Obstet Gynecol 1989; 74:930-933.[Medline]
7. Hricak H, Stern JL, Fisher MR, Shapeero LG, Winkler ML, Lacey CG. Endometrial carcinoma staging by MR imaging. Radiology 1987; 162:297-305.[Abstract/Free Full Text]
8. Javitt MC, Stein HL, Lovecchio JL. MRI in staging of endometrial and cervical carcinoma. Magn Reson Imaging 1987; 5:83-92.[Medline]
9. Sironi S, Mellone R, Venzulli A, et al. Assessment of the myometrial infiltration of endometrial carcinoma (FIGO stage I–II): the accuracy of magnetic resonance. Radiol Med 1989; 77:386-390.
10. Yazigi R, Cohen G, Munoz AK, et al. Magnetic resonance imaging determination of myometrial invasion in endometrial carcinoma. Gynecol Oncol 1989; 34:94-97.[Medline]
11. Hacker NF. Uterine cancer. In: Berek JS, Hacker NF, eds. Practical gynecologic oncology. Baltimore, Md: Williams & Wilkins, 1989; 285-326.
12. Lave JR, Pashos CL, Anderson GF, et al. Costing medical care: using medicare administrative data. Med Care 1994; 32:JS77-JS89.[Medline]
13. Sironi S, Colombo E, Villa G, et al. Myometrial invasion by endometrial carcinoma: assessment with plain and gadolinium-enhanced MR imaging. Radiology 1992; 185:207-212.[Abstract/Free Full Text]
14. Yamashita Y, Mizutani H, Torashima M, et al. Assessment of myometrial invasion by endometrial carcinoma: transvaginal sonography versus contrast-enhanced MR imaging. AJR Am J Roentgenol 1993; 161:595-599.[Abstract/Free Full Text]
15. DelMaschio A, Vanzulli A, Sironi S, et al. Estimating the depth of myometrial involvement by endometrial carcinoma: efficacy of transvaginal sonography versus MR imaging. AJR Am J Roentgenol 1993; 160:533-538.[Abstract/Free Full Text]
16. Sironi S, Taccagni G, Garancini P, et al. Myometrial invasion by endometrial carcinoma: assessment by MR imaging. AJR Am J Roentgenol 1992; 158:565-569.[Abstract/Free Full Text]
17. Gordon AN, Fleisher AC, Dudley BS, et al. Preoperative assessment of myometrial invasion of endometrial adenocarcinoma by sonography (US) and magnetic resonance imaging (MRI). Gynecol Oncol 1989; 34:175-179.[Medline]

This article has been cited by other articles:

				P. Torricelli, S. Ferraresi, F. Fiocchi, G. Ligabue, V. M. Jasonni, I. Di Monte, and F. Rivasi 3-T MRI in the Preoperative Evaluation of Depth of Myometrial Infiltration in Endometrial Cancer Am. J. Roentgenol., February 1, 2008; 190(2): 489 - 495. [Abstract] [Full Text] [PDF]

				J M McHugo and J Olliff Endometrial cancer Imaging, August 1, 2006; 18(1): 1 - 9. [Abstract] [Full Text] [PDF]

				R. Manfredi, P. Mirk, G. Maresca, P. A. Margariti, A. Testa, G. F. Zannoni, D. Giordano, G. Scambia, and P. Marano Local-Regional Staging of Endometrial Carcinoma: Role of MR Imaging in Surgical Planning Radiology, May 1, 2004; 231(2): 372 - 378. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hardesty, L. A. Articles by Kelley, J. L. Search for Related Content PubMed PubMed Citation Articles by Hardesty, L. A. Articles by Kelley, J. L.