Ann Surg Oncol DOI 10.1245/s10434-013-3479-3
ORIGINAL ARTICLE – ENDOCRINE TUMORS
Operative Failure in Minimally Invasive Parathyroidectomy Utilizing an Intraoperative Parathyroid Hormone Assay Sukhyung Lee, MD1,3, Haengrang Ryu, MD1,4, Lilah F. Morris, MD1, Elizabeth G. Grubbs, MD1, Jeffrey E. Lee, MD1, Nusrat Harun, MS2, Lei Feng, MS2, and Nancy D. Perrier, MD1 1
Section of Surgical Endocrinology, Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; 2Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX; 3 Department of Surgery, John P. Murtha Cancer Center, Walter Reed National Military Medical Center, Bethesda, MD; 4 Department of Surgery, Hongik Hospital, Seoul, Republic of Korea
ABSTRACT Background. Minimally invasive parathyroidectomy (MIP) is a targeted operation to cure primary hyperparathyroidism utilizing intraoperative parathyroid hormone monitoring (IOPTH). The purpose of this study was to quantify the operative failure of MIP. Methods. Utilizing institutional parathyroid surgery database, demographic, operative, and biochemical data were analyzed for successful and failed MIP. Operative failure was defined as \6 months of eucalcemia after operation. Results. Five hundred thirty-eight patients (96.6 %) had successful MIP with mean follow-up of 13 months, and 19 (3.4 %) had operative failure. The major cause of operative failure (11 of 19) was the result of surgeons’ inability to identify all abnormal parathyroid glands. The remaining eight operative failures were the result of falsely positive IOPTH results. Eleven of 19 patients whose MIP had failed underwent a second parathyroid surgery. All but one of these patients achieved operative success, and 9 patients had missed multigland disease. Only 46 (8.3 %) of 557 patients had conversion to bilateral cervical exploration (BCE). Eighty percent of patients had more than 70 % IOPTH decrease, and all had successful operations. Patients with a marginal IOPTH decrease (50–59 %) had a treatment failure rate of 20 %. Conclusions. The most common cause of operative failure in MIP utilizing IOPTH was the result of surgeons’ failure to identify all abnormal parathyroid glands. Falsely positive IOPTH is rare, and a targeted MIP utilizing IOPTH can
Ó Society of Surgical Oncology 2014 First Received: 8 July 2013 S. Lee, MD e-mail: [email protected]
achieve an excellent operative success rate without routine BCE. Selective BCE on patients with marginal IOPTH decrease may improve surgical outcome.
Primary hyperparathyroidism (PHPT) is the most common cause of hypercalcemia. If untreated, hyperparathyroidism can increase the risk of cardiac disease, osteoporosis, and kidney disease.1,2 Hyperparathyroidism is also associated with subjective complaints of memory loss, sleep disturbance, and other nonspecific symptoms that decrease quality of life.3,4 Surgery is currently the only curative therapy for PHPT. Historically, bilateral cervical exploration (BCE) examining all four parathyroid glands was performed with general endotracheal anesthesia, extended neck exploration, and overnight monitoring. In recent decades, improved localization studies and the availability of intraoperative parathyroid hormone monitoring (IOPTH) assays have made minimally invasive parathyroidectomy (MIP) feasible, and this technique has rapidly gained popularity as an initial treatment choice for PHPT.5–8 More than 80 % of patients with PHPT have singlegland disease, which can be treated using MIP guided by IOPTH with minimal neck dissection in an outpatient setting. In patients with multigland disease (MGD), some surgeons argue that IOPTH is inaccurate for determining complete excision. These surgeons advocate for routine BCE with identification of all parathyroid glands and excision of all enlarged glands.9,10 Despite excellent success rates for MIP guided by IOPTH, operative failures do occur. The purpose of this study was to describe the causes of operative failure in MIP guided by IOPTH with preoperative imaging at a single specialized care center.
S. Lee et al.
METHODS Patient Selection This study was approved by The University of Texas MD Anderson Cancer Center Institutional Review Board. We evaluated our institution’s prospectively maintained surgical endocrinology database that contains data on 1,243 patients who underwent parathyroidectomy from 1998 to 2010. From this database, 877 patients who were diagnosed with sporadic PHPT, and underwent initial MIP were identified. 557 patients who had a minimum of 6 months of follow-up data or considered operative failures are the subject of this study. The remaining 320 patients, even though eucalcemic postoperatively, were excluded because of the insufficient length of their follow-up (less than 6 months). Patients were also excluded if they had undergone any previous parathyroid (113 patients) or thyroid surgical procedure (87 patients) or had familial or multiple endocrine neoplasia-related hyperparathyroidism syndrome (87 patients). In addition, patients with secondary or tertiary hyperparathyroidism were excluded (22 patients). Patients were also excluded if the intended initial operation was a planned BCE, such as when the disease could not be localized using preoperative imaging (56 patients). Ten patients with parathyroid carcinoma and one patient with lithium associated hyperparathyroidism were also excluded.
more but hypercalcemia persisted after the surgical intervention. Operative success was defined as documented eucalcemia 6 months after parathyroidectomy, regardless of follow-up PTH result. Operative failure (disease persistence) was defined as hypercalcemia with inappropriately elevated PTH within 6 months after parathyroidectomy. In addition, other causes of hypercalcemia such as malignancy, sarcoidosis, or medication use were evaluated and ruled out as the cause of hypercalcemia in operative failure cases. MGD was defined as present when more than one abnormal gland (hyperplastic or adenoma on final pathology) was removed at the time of the first operation or when excision of single pathologically abnormal parathyroid gland did not result in operative success. Data Analysis We calculated descriptive statistics for the entire patient cohort and for the operative success (cure) and operative failure subgroups. Fisher’s exact test was used to examine associations between outcome (operative success or failure) and various dichotomous or categorical variables. The Wilcoxon rank-sum test was used for continuous variables. Statistical analyses were performed using commercially available software (SAS version 9.3; SAS Institute, Cary, NC, USA) by our Department of Biostatistics (NH and LF). P \ 0.05 was considered statistically significant.
Data Collection RESULTS For each patient included, patient demographics; pre-, intra-, and post-operative biochemical data; initial and subsequent surgical procedures; and outcomes from the database were recorded. Surgical Procedure In all cases, the initial dissection was performed in accordance with the findings noted on the preoperative radiological localization studies. Studies included one or a combination of technetium 99mTc sestamibi imaging, ultrasonography, or four-dimensional computed tomographic scanning. Termination of the operation was based on the clinical decision of the operating surgeon, IOPTH results, or both. The intraoperative criterion for successful surgery was a decrease of intact parathyroid hormone (PTH) levels 50 % or more from the preincision value in a peripheral blood sample obtained 10 min after removal of all evident abnormal parathyroid tissue. In general, BCE was performed if there was a clinical suspicion of MGD or failure to achieve a 50 % or more decrease in IOPTH levels. IOPTH test results were considered falsely positive if the IOPTH decrease was 50 % or
Among 557 patients who underwent MIP as an initial surgical procedure and who had known operative failure or follow-up data for more than 6 months after surgery, 538 patients (96.6 %) underwent successful surgery, as confirmed by sustained eucalcemia for 6 months (median follow-up, 13 months). Nineteen patients (3.4 %) experienced operative failure. Patients who experienced operative failure had higher preoperative serum calcium and creatinine levels (Table 1). Age, sex, race, body mass index, and preoperative PTH levels were not significantly different between the patients who had successful surgeries and those who did not. The leading cause of operative failure (11 of 19 operative failures) was the surgeon’s failure to remove all abnormal parathyroid glands. Of those 11 failed operative interventions, 7 failures were the result of terminating the procedure without performing BCE with inadequate IOPTH decreases, and 4 failures were the result of inability to identify all abnormal glands even with BCE. The remaining 8 operative failures were the result of falsely positive IOPTH results. Eleven of 19 patients whose parathyroid surgery failed underwent a second parathyroid
Operative Failure in Minimally Invasive Parathyroidectomy TABLE 1 Demographic, preoperative, and postoperative biochemical and surgical characteristics Characteristic
Operative Operative success failure (n = 538) (n = 19)
Age, mean (SD), year
60.5 (12.6) 64.5 (12.7)
Sex, n (%) 117 (21.7) 1 (5.3)
Female
421 (78.3) 18 (94.7)
Ethnicity, n (%) Nonwhite BMI, kg/m2, mean ± SD
0.282 0.093
Male
White
P value
0.753 447 (83.1) 17 (89.5) 91 (16.9)
2 (10.5)
28.5 ± 6.9 27.9 ± 6.6
0.696
Preoperative laboratory values, mean ± SD Calcium, mg/mL
10.7 ± 0.7 11.1 ± 0.7
0.01
PTH, pg/mL
145 ± 163 144 ± 76
0.459
Creatinine, mg/dL
0.9 ± 0.3
1.3 ± 0.6
Side of surgery, n (%) Right Left Bilateral No. of glands removed, mean ± SD
0.004 0.206
250 (46.4) 6 (31.6) 245 (45.4) 10 (52.6) 43 (8.2)
3 (15.8)
0.309
1.2 ± 0.5
1.3 ± 0.7
0.196
Glands removed, n (%)
0.027
Single
465 (86.4) 13 (68.4)
Multiple
73 (14.1)
6 (31.6)
Intraoperative laboratory values Preincision PTH, mean ± SD
227 ± 366 182 ± 102
10 min PTH, mean ± SD
39 ± 36
0.849
122 ± 71
\0.001
31 ± 25
\0.001
IOPTH decrease C50 %, n (%) 514 (95.6) 8 (42.1) Postoperative laboratory values, mean ± SD
\0.001
IOPTH decrease, %, mean ± SD 79 ± 12
Calcium, mg/dL
9.4 ± 0.4
10.7 ± 0.4 \0.001
PTH, pg/mL
52 ± 29
111 ± 55
\0.001
SD standard deviation, BMI body mass index, IOPTH intraoperative parathyroid hormone, PTH parathyroid hormone
operation. All patients but 1 who underwent a second operation achieved operative success. The patient who underwent the second operation but failed to achieve operative success had subtotal parathyroidectomy with a 57 % IOPTH decrease. Eight patients did not undergo a second operations for various reasons (medical comorbidities or patient choice). Of the 11 patients who underwent a second operation, 9 had MGD on final pathologic review. One patient was found to have initially had an incomplete excision of a single involved gland. One patient was cured after a second operation even though no parathyroid gland was identified in the second operation (Table 2). Among the entire cohort of 557 patients, 46 patients’ (8.3 %) procedures were converted to BCE. The rates of BCE were similar in the operative success and operative
failure groups. Eight patients underwent BCE because of a failed localization study (normal parathyroid glands were identified in the expected side of the initial exploration), and 1 patient had MGD on final pathological evaluation. Six patients underwent BCE as a result of an intraoperative clinical suspicion of MGD on the basis of surgeons’ visual evaluation of the gland size (both parathyroid glands were abnormal on the explored side). Of these, all 6 patients had MGD on pathology. 32 patients underwent BCE as a result of inadequate IOPTH value reduction. Three of 46 patients who underwent BCE had a failed operation; 43 of 46 had operative success. IOPTH data were used in 547 patients. Even though IOPTH was intended to be used in all 557 patients, IOPTH was not available in 10 patients as the result of technical/ laboratory issues. When the IOPTH value was categorized by extent of decline, 11 (42 %) of 26 patients who had less than 50 % IOPTH decrease had persistent disease. Six (20 %) of 30 patients who had a 50–59 % IOPTH decrease experienced surgical failure. 69 patients had a 60–69 % IOPTH decrease, and of these, only 2 (3 %) experienced a failure to cure. 422 patients had a 70 % or greater decline of IOPTH level with no failure (Table 3). Different IOPTH criteria for operative termination are compared in Table 4, and sensitivity, specificity, positive predictive values, negative predictive values, and accuracies are provided in Table 5. Utilization of IOPTH decrease criteria of greater than 50 % can decrease the failure rate but increases the rate of nontherapeutic BCE. If we increase the IOPTH decrease level for the termination of surgery to 60 %, 6 operative failures could have been prevented, with 24 patients undergoing potentially unnecessary BCE. Accuracies of operative success utilizing 50 % IOPTH decrease, 60 % IOPTH decrease, and 50 % IOPTH decrease with normalized IOPTH level were 95.8, 92.5, and 90.7 %, respectively. An IOPTH decrease of 50 % or more still maintained the highest accuracy in our comparison (Table 5). Six of the cured patients had recurrent hyperparathyroidism. Median time from their initial surgery to recurrence was 3 years.
DISCUSSION MIP guided by IOPTH performed in patients with sporadic hyperparathyroidism had an operative success rate of 96.6 % with less than 10 % BCE. Ten additional patients achieved operative success after a second operation, for an overall success rate of 98.4 % (546 of 557). The most common cause of surgical failure (11 of 19) in our study was missed MGD by operating surgeons. Falsely positive IOPTH results were responsible for the remaining 8 surgical failures.
S. Lee et al. TABLE 2 Operative details of 19 patients who failed the initial MIP and outcome of the second operation Patient no. Reason for failure
IOPTH Initial decrease surgery (%)
Quadrant Second from operation/reason which glands for no surgery removed
1
Fail to identify gland
0
MIP to BCE LS
BCE
None
Unable to identify S
2
Fail to identify gland
13
MIP to BCE LS
BCE
RI
Multigland disease
S
3
Fail to identify gland
30
MIP to BCE LS, RS, LI
BCE
LI
Multigland disease
S
4
Surgeon decision 3
MIP
LS, LI
MIP
RS, 1/2RI
Multigland disease
S
5
Surgeon decision 25
MIP
LS
BCE
RI, RS
Multigland disease
S
6
Surgeon decision 39
MIP
RS
BCE
LS, LI
Multigland disease
S
7
Surgeon decision 3
MIP
RI
BCE
LI, 1/2LS
Multigland disease
S
8
Surgeon decision 0
MIP
LI
MIP
LI
Incomplete excision
S
9
False IOPTH
53
MIP
LS
BCE
LI
Multigland disease
S
10
False IOPTH
61
MIP
LI
MIP
RI
Multigland disease
S
11
False IOPTH
57
MIP
RS, RI
MIP
LI, 1/2LS
Multigland disease
F
12
Fail to Identify gland
0
MIP
Medical management
13
Surgeon decision 0
MIP
Medical reason
14
Surgeon decision 23
MIP
Surgery planned
15
False IOPTH
55
MIP
Medical reason
16
False IOPTH
56
MIP
Medical reason
17
False IOPTH
64
MIP
No longer met criteria
18
False IOPTH
50
MIP
No longer met criteria
19
False IOPTH
58
MIP
Patient did not follow up
Quadrant from Operative finding which glands removed, second operation
Outcome
IOPTH intraoperative parathyroid hormone, MIP minimally invasive parathyroidectomy, BCE bilateral cervical exploration, LS left superior parathyroid, LI left inferior parathyroid, RS right superior parathyroid, RI right inferior parathyroid, S operative success, F operative failure
Of the 11 operative failures that were the result of missed MGD by surgeons, 7 cases were due to surgeon’s operative decision making by terminating the operation with an inadequate decline in IOPTH and not performing BCE. Clinical decision making for early termination for these 7 patients included removal of radiographically concordant suspected single-gland disease with identification of a normal ipsilateral second parathyroid gland. Other factors (such as a desire to minimize anesthetic time) also led to termination of the procedure. Because the IOPTH assays are performed in the laboratory, which is remote from the operating theater, there is a delay. Of these 7 patients, 5 underwent second operations. Four patients had unrecognized MGD in the initial operation, and 1 patient
had incomplete excision of a dumbbell-shaped, single abnormal parathyroid gland. The remaining 4 operative failures were the result of surgeons’ combined skill and judgment. These 4 patients underwent BCE, and surgeon failed to identify all abnormal parathyroid glands. Lew et al. reviewed a total of 723 patients with 21 operative failures (2.9 %). The main cause of their operative failures was an inability to find the abnormal parathyroid glands (16 of 21 patients, 76 %).11 Our study also confirms their finding that surgeons’ clinical skills and judgment play an important role in the surgical outcome of MIP. The remainder of the operative failures (8 of 19) were the result of falsely positive IOPTH results. IOPTH decreased 50 % or more without operative success in these
Operative Failure in Minimally Invasive Parathyroidectomy
cases. In the present study, we found that the percentage decrease was important because patients with robust PTH declines were more likely to achieve operative success. We found that 80 % of patients will have greater than 70 % IOPTH decrease with 0 % failure rate, whereas a decrease of 50–59 % had a failure rate of 20 %. Adding the additional criterion of PTH levels falling into the normal range in addition to IOPTH decrease of 50 % or more did not seem to improve surgical outcome significantly in our study (Tables 4, 5). Increasing the IOPTH decrease level to 60 % or more also did not improve the accuracy of the test (Table 5). Previous studies have shown that stricter criteria TABLE 3 IOPTH decrease and operative outcome Level of IOPTH decrease (%) Operative success Operative failure (n = 528) (n = 19) \50
15 (57.7 %)
11 (42.3 %)
50–59
24 (80.0 %)
6 (20.0 %)
60–69
67 (97.1 %)
2 (2.9 %)
C70
422 (100.0 %)
0 (0.0 %)
IOPTH intraoperative parathyroid hormone
TABLE 4 Comparison of IOPTH criteria IOPTH status
Operative success n = 538 (96.6 %)
Operative failure n = 19 (3.4 %)
Yes
513 (97.2)
8 (42.1)
No
15 (2.8)
11 (57.9)
P value
\0.001
IOPTH decrease C50 %
\0.001
IOPTH decrease C60 % Yes
489 (92.6)
2 (10.5)
No
39 (7.4)
17 (89.5) \0.001
IOPTH decrease C50 % and PTH B80 Yes
482 (91.3)
5 (26.3)
No
46 (8.7)
14 (73.7)
IOPTH intraoperative parathyroid hormone
do not significantly improve the operation’s success rate.12 However, with a failure rate close to 20 % in the group with a lower IOPTH decrease (50–59 %), selective BCE can be considered in patients with marginal IOPTH decrease. Among the cohort of patients who underwent a BCE, the decision to convert to bilateral surgery was prompted by IOPTH in 67 % (31 of 46) of the patients. In all of these cases, unsuspected additional abnormal parathyroid glands were identified. Other patients underwent BCE on the basis of the surgeon’s clinical suspicion of MGD, such as normal parathyroid gland by surgeon visual inspection on the explored side. Previous studies showed that surgical experience plays a critical role in the overall success of parathyroid surgery.13 We believe that the surgeon’s experience, augmented by IOPTH, can decrease the need for unselective BCE while still increasing the identification of abnormal parathyroid gland. Routine BCE would not improve the successful outcome significantly because the false-positive rate of IOPTH is very low (1.4 %, 8 of 557). Even though BCE can be performed with minimal complications in experienced hands, complications such as recurrent laryngeal nerve injury and hypoparathyroidism do occur. Patients who had MIP only put one side of the neck structure at risk. The potential benefit of performing MIP is that the contralateral dissection can be performed in the future in a region without scarring or an increased risk of nerve injury, unlike a procedure being performed after initial BCE would. Patients in our study who had operative failure had higher mean preoperative serum calcium and creatinine levels. However, it is difficult to judge the clinical significance of these findings. An element of secondary hyperparathyroidism and/or higher prevalence of parathyroid hyperplasia or MGD may be responsible. Age, sex, race, body mass index, and preoperative PTH levels did not differ significantly between the patients who had successful surgeries and those whose surgeries failed. This study has several limitations. As a retrospective descriptive study, hypotheses that test comparisons of MIP to BCE cannot be performed. Fortunately, operative failure
TABLE 5 IOPTH accuracy rates with different IOPTH criteria IOPTH decrease
Sensitivity (TP/TP ? FN)
Specificity (TN/TN ? FP)
PPV (TP/TP ? FP)
NPV (TN/TN ? FN)
Accuracy (TP ? TN/ TP ? TN ? FP ? FN)
C50 %
97.2
57.9
98.5
42.3
95.8
C60 %
92.6
89.5
99.6
30.4
92.5
C50 % and PTH B80
91.3
73.7
99.0
23.3
90.7
IOPTH intraoperative parathyroid hormone, TP truly positive—correct prediction of postoperative cure, TN truly negative—correct prediction of postoperative failure, FP falsely positive—incorrect prediction of postoperative cure, FN falsely negative—incorrect prediction of postoperative failure, PPV positive predictive value, NPV negative predictive value
S. Lee et al.
after parathyroid surgery is not common. To better understand the cause of failure, a large number of patients who undergo parathyroidectomy is needed. Therefore, even though this study is descriptive in nature, it can contribute the better understanding of the reason for failure in MIP. Second, our follow-up data are suboptimal. More than 30 % of our patients did not complete 6-month follow-up, likely as a result of a geographically diverse referral base. All patients who did not complete 6-month follow-up, however, had documented normal calcium level postoperatively. In conclusion, a targeted MIP guided by preoperative radiological localization and IOPTH can achieve an excellent operative success rate without the need for routine BCE. The most common cause of operative failure of MIP in our study was surgeons’ inability to identify all abnormal parathyroid glands. Patients who have more than 70 % IOPTH decrease are likely cured and do not need BCE. When there is an IOPTH decrease of more than 50 % but less than 60 %, the surgeon should use his or her judgment as to whether the patient needs BCE. ACKNOWLEDGMENT Supported in part by The MD Anderson Cancer Center Support Grant CA016672. We thank Zach Bohannan and Melissa Burkett from The University of Texas MD Anderson Cancer Center Department of Scientific Publications for their assistance. DISCLOSURE
The authors declare no conflict of interest.
REFERENCES 1. Piovesan A, Molineri N, Casasso F, et al. Left ventricular hypertrophy in primary hyperparathyroidism. Effects of successful parathyroidectomy. Clin Endocrinol (Oxf). 1999;50:321–8.
2. Rubin MR, Bilezikian JP, McMahon DJ, et al. The natural history of primary hyperparathyroidism with or without parathyroid surgery after 15 years. J Clin Endocrinol Metab. 2008;93:3462– 70. 3. Edwards ME, Rotramel A, Beyer T, et al. Improvement in the health-related quality-of-life symptoms of hyperparathyroidism is durable on long-term follow-up. Surgery. 2006;140:655–63. 4. Espiritu RP, Kearns AE, Vickers KS, Grant C, Ryu E, Wermers RA. Depression in primary hyperparathyroidism: prevalence and benefit of surgery. J Clin Endocrinol Metab. 2011;96:E1737–45. 5. Irvin GL 3rd, Carneiro DM. Management changes in primary hyperparathyroidism. JAMA. 2000;284:934–6. 6. Irvin GL 3rd, Carneiro DM, Solorzano CC. Progress in the operative management of sporadic primary hyperparathyroidism over 34 years. Ann Surg. 2004;239:704–8. 7. Udelsman R, Lin Z, Donovan P. The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism. Ann Surg. 2011;253:585–91. 8. Greene AB, Butler RS, McIntyre S, et al. National trends in parathyroid surgery from 1998 to 2008: a decade of change. J Am Coll Surg. 2009;209:332–43. 9. Norman J, Lopez J, Politz D. Abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid operations. J Am Coll Surg. 2012;214:260–9. 10. Siperstein A, Berber E, Barbosa GF, et al. Predicting the success of limited exploration for primary hyperparathyroidism using ultrasound, sestamibi, and intraoperative parathyroid hormone: analysis of 1158 cases. Ann Surg. 2008;248:420–8. 11. Lew JI, Rivera M, Irvin GL 3rd, Solorzano CC. Operative failure in the era of focused parathyroidectomy: a contemporary series of 845 patients. Arch Surg. 2010;145:628–33. 12. Barczynski M, Konturek A, Hubalewska-Dydejczyk A, Cichon S, Nowak W. Evaluation of Halle, Miami, Rome, and Vienna intraoperative iPTH assay criteria in guiding minimally invasive parathyroidectomy. Langenbecks Arch Surg. 2009;394:843–9. 13. Chen H, Wang TS, Yen TW, et al. Operative failures after parathyroidectomy for hyperparathyroidism: the influence of surgical volume. Ann Surg. 2010;252:691–5.
ORIGINAL ARTICLE – ENDOCRINE TUMORS
Operative Failure in Minimally Invasive Parathyroidectomy Utilizing an Intraoperative Parathyroid Hormone Assay Sukhyung Lee, MD1,3, Haengrang Ryu, MD1,4, Lilah F. Morris, MD1, Elizabeth G. Grubbs, MD1, Jeffrey E. Lee, MD1, Nusrat Harun, MS2, Lei Feng, MS2, and Nancy D. Perrier, MD1 1
Section of Surgical Endocrinology, Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; 2Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX; 3 Department of Surgery, John P. Murtha Cancer Center, Walter Reed National Military Medical Center, Bethesda, MD; 4 Department of Surgery, Hongik Hospital, Seoul, Republic of Korea
ABSTRACT Background. Minimally invasive parathyroidectomy (MIP) is a targeted operation to cure primary hyperparathyroidism utilizing intraoperative parathyroid hormone monitoring (IOPTH). The purpose of this study was to quantify the operative failure of MIP. Methods. Utilizing institutional parathyroid surgery database, demographic, operative, and biochemical data were analyzed for successful and failed MIP. Operative failure was defined as \6 months of eucalcemia after operation. Results. Five hundred thirty-eight patients (96.6 %) had successful MIP with mean follow-up of 13 months, and 19 (3.4 %) had operative failure. The major cause of operative failure (11 of 19) was the result of surgeons’ inability to identify all abnormal parathyroid glands. The remaining eight operative failures were the result of falsely positive IOPTH results. Eleven of 19 patients whose MIP had failed underwent a second parathyroid surgery. All but one of these patients achieved operative success, and 9 patients had missed multigland disease. Only 46 (8.3 %) of 557 patients had conversion to bilateral cervical exploration (BCE). Eighty percent of patients had more than 70 % IOPTH decrease, and all had successful operations. Patients with a marginal IOPTH decrease (50–59 %) had a treatment failure rate of 20 %. Conclusions. The most common cause of operative failure in MIP utilizing IOPTH was the result of surgeons’ failure to identify all abnormal parathyroid glands. Falsely positive IOPTH is rare, and a targeted MIP utilizing IOPTH can
Ó Society of Surgical Oncology 2014 First Received: 8 July 2013 S. Lee, MD e-mail: [email protected]
achieve an excellent operative success rate without routine BCE. Selective BCE on patients with marginal IOPTH decrease may improve surgical outcome.
Primary hyperparathyroidism (PHPT) is the most common cause of hypercalcemia. If untreated, hyperparathyroidism can increase the risk of cardiac disease, osteoporosis, and kidney disease.1,2 Hyperparathyroidism is also associated with subjective complaints of memory loss, sleep disturbance, and other nonspecific symptoms that decrease quality of life.3,4 Surgery is currently the only curative therapy for PHPT. Historically, bilateral cervical exploration (BCE) examining all four parathyroid glands was performed with general endotracheal anesthesia, extended neck exploration, and overnight monitoring. In recent decades, improved localization studies and the availability of intraoperative parathyroid hormone monitoring (IOPTH) assays have made minimally invasive parathyroidectomy (MIP) feasible, and this technique has rapidly gained popularity as an initial treatment choice for PHPT.5–8 More than 80 % of patients with PHPT have singlegland disease, which can be treated using MIP guided by IOPTH with minimal neck dissection in an outpatient setting. In patients with multigland disease (MGD), some surgeons argue that IOPTH is inaccurate for determining complete excision. These surgeons advocate for routine BCE with identification of all parathyroid glands and excision of all enlarged glands.9,10 Despite excellent success rates for MIP guided by IOPTH, operative failures do occur. The purpose of this study was to describe the causes of operative failure in MIP guided by IOPTH with preoperative imaging at a single specialized care center.
S. Lee et al.
METHODS Patient Selection This study was approved by The University of Texas MD Anderson Cancer Center Institutional Review Board. We evaluated our institution’s prospectively maintained surgical endocrinology database that contains data on 1,243 patients who underwent parathyroidectomy from 1998 to 2010. From this database, 877 patients who were diagnosed with sporadic PHPT, and underwent initial MIP were identified. 557 patients who had a minimum of 6 months of follow-up data or considered operative failures are the subject of this study. The remaining 320 patients, even though eucalcemic postoperatively, were excluded because of the insufficient length of their follow-up (less than 6 months). Patients were also excluded if they had undergone any previous parathyroid (113 patients) or thyroid surgical procedure (87 patients) or had familial or multiple endocrine neoplasia-related hyperparathyroidism syndrome (87 patients). In addition, patients with secondary or tertiary hyperparathyroidism were excluded (22 patients). Patients were also excluded if the intended initial operation was a planned BCE, such as when the disease could not be localized using preoperative imaging (56 patients). Ten patients with parathyroid carcinoma and one patient with lithium associated hyperparathyroidism were also excluded.
more but hypercalcemia persisted after the surgical intervention. Operative success was defined as documented eucalcemia 6 months after parathyroidectomy, regardless of follow-up PTH result. Operative failure (disease persistence) was defined as hypercalcemia with inappropriately elevated PTH within 6 months after parathyroidectomy. In addition, other causes of hypercalcemia such as malignancy, sarcoidosis, or medication use were evaluated and ruled out as the cause of hypercalcemia in operative failure cases. MGD was defined as present when more than one abnormal gland (hyperplastic or adenoma on final pathology) was removed at the time of the first operation or when excision of single pathologically abnormal parathyroid gland did not result in operative success. Data Analysis We calculated descriptive statistics for the entire patient cohort and for the operative success (cure) and operative failure subgroups. Fisher’s exact test was used to examine associations between outcome (operative success or failure) and various dichotomous or categorical variables. The Wilcoxon rank-sum test was used for continuous variables. Statistical analyses were performed using commercially available software (SAS version 9.3; SAS Institute, Cary, NC, USA) by our Department of Biostatistics (NH and LF). P \ 0.05 was considered statistically significant.
Data Collection RESULTS For each patient included, patient demographics; pre-, intra-, and post-operative biochemical data; initial and subsequent surgical procedures; and outcomes from the database were recorded. Surgical Procedure In all cases, the initial dissection was performed in accordance with the findings noted on the preoperative radiological localization studies. Studies included one or a combination of technetium 99mTc sestamibi imaging, ultrasonography, or four-dimensional computed tomographic scanning. Termination of the operation was based on the clinical decision of the operating surgeon, IOPTH results, or both. The intraoperative criterion for successful surgery was a decrease of intact parathyroid hormone (PTH) levels 50 % or more from the preincision value in a peripheral blood sample obtained 10 min after removal of all evident abnormal parathyroid tissue. In general, BCE was performed if there was a clinical suspicion of MGD or failure to achieve a 50 % or more decrease in IOPTH levels. IOPTH test results were considered falsely positive if the IOPTH decrease was 50 % or
Among 557 patients who underwent MIP as an initial surgical procedure and who had known operative failure or follow-up data for more than 6 months after surgery, 538 patients (96.6 %) underwent successful surgery, as confirmed by sustained eucalcemia for 6 months (median follow-up, 13 months). Nineteen patients (3.4 %) experienced operative failure. Patients who experienced operative failure had higher preoperative serum calcium and creatinine levels (Table 1). Age, sex, race, body mass index, and preoperative PTH levels were not significantly different between the patients who had successful surgeries and those who did not. The leading cause of operative failure (11 of 19 operative failures) was the surgeon’s failure to remove all abnormal parathyroid glands. Of those 11 failed operative interventions, 7 failures were the result of terminating the procedure without performing BCE with inadequate IOPTH decreases, and 4 failures were the result of inability to identify all abnormal glands even with BCE. The remaining 8 operative failures were the result of falsely positive IOPTH results. Eleven of 19 patients whose parathyroid surgery failed underwent a second parathyroid
Operative Failure in Minimally Invasive Parathyroidectomy TABLE 1 Demographic, preoperative, and postoperative biochemical and surgical characteristics Characteristic
Operative Operative success failure (n = 538) (n = 19)
Age, mean (SD), year
60.5 (12.6) 64.5 (12.7)
Sex, n (%) 117 (21.7) 1 (5.3)
Female
421 (78.3) 18 (94.7)
Ethnicity, n (%) Nonwhite BMI, kg/m2, mean ± SD
0.282 0.093
Male
White
P value
0.753 447 (83.1) 17 (89.5) 91 (16.9)
2 (10.5)
28.5 ± 6.9 27.9 ± 6.6
0.696
Preoperative laboratory values, mean ± SD Calcium, mg/mL
10.7 ± 0.7 11.1 ± 0.7
0.01
PTH, pg/mL
145 ± 163 144 ± 76
0.459
Creatinine, mg/dL
0.9 ± 0.3
1.3 ± 0.6
Side of surgery, n (%) Right Left Bilateral No. of glands removed, mean ± SD
0.004 0.206
250 (46.4) 6 (31.6) 245 (45.4) 10 (52.6) 43 (8.2)
3 (15.8)
0.309
1.2 ± 0.5
1.3 ± 0.7
0.196
Glands removed, n (%)
0.027
Single
465 (86.4) 13 (68.4)
Multiple
73 (14.1)
6 (31.6)
Intraoperative laboratory values Preincision PTH, mean ± SD
227 ± 366 182 ± 102
10 min PTH, mean ± SD
39 ± 36
0.849
122 ± 71
\0.001
31 ± 25
\0.001
IOPTH decrease C50 %, n (%) 514 (95.6) 8 (42.1) Postoperative laboratory values, mean ± SD
\0.001
IOPTH decrease, %, mean ± SD 79 ± 12
Calcium, mg/dL
9.4 ± 0.4
10.7 ± 0.4 \0.001
PTH, pg/mL
52 ± 29
111 ± 55
\0.001
SD standard deviation, BMI body mass index, IOPTH intraoperative parathyroid hormone, PTH parathyroid hormone
operation. All patients but 1 who underwent a second operation achieved operative success. The patient who underwent the second operation but failed to achieve operative success had subtotal parathyroidectomy with a 57 % IOPTH decrease. Eight patients did not undergo a second operations for various reasons (medical comorbidities or patient choice). Of the 11 patients who underwent a second operation, 9 had MGD on final pathologic review. One patient was found to have initially had an incomplete excision of a single involved gland. One patient was cured after a second operation even though no parathyroid gland was identified in the second operation (Table 2). Among the entire cohort of 557 patients, 46 patients’ (8.3 %) procedures were converted to BCE. The rates of BCE were similar in the operative success and operative
failure groups. Eight patients underwent BCE because of a failed localization study (normal parathyroid glands were identified in the expected side of the initial exploration), and 1 patient had MGD on final pathological evaluation. Six patients underwent BCE as a result of an intraoperative clinical suspicion of MGD on the basis of surgeons’ visual evaluation of the gland size (both parathyroid glands were abnormal on the explored side). Of these, all 6 patients had MGD on pathology. 32 patients underwent BCE as a result of inadequate IOPTH value reduction. Three of 46 patients who underwent BCE had a failed operation; 43 of 46 had operative success. IOPTH data were used in 547 patients. Even though IOPTH was intended to be used in all 557 patients, IOPTH was not available in 10 patients as the result of technical/ laboratory issues. When the IOPTH value was categorized by extent of decline, 11 (42 %) of 26 patients who had less than 50 % IOPTH decrease had persistent disease. Six (20 %) of 30 patients who had a 50–59 % IOPTH decrease experienced surgical failure. 69 patients had a 60–69 % IOPTH decrease, and of these, only 2 (3 %) experienced a failure to cure. 422 patients had a 70 % or greater decline of IOPTH level with no failure (Table 3). Different IOPTH criteria for operative termination are compared in Table 4, and sensitivity, specificity, positive predictive values, negative predictive values, and accuracies are provided in Table 5. Utilization of IOPTH decrease criteria of greater than 50 % can decrease the failure rate but increases the rate of nontherapeutic BCE. If we increase the IOPTH decrease level for the termination of surgery to 60 %, 6 operative failures could have been prevented, with 24 patients undergoing potentially unnecessary BCE. Accuracies of operative success utilizing 50 % IOPTH decrease, 60 % IOPTH decrease, and 50 % IOPTH decrease with normalized IOPTH level were 95.8, 92.5, and 90.7 %, respectively. An IOPTH decrease of 50 % or more still maintained the highest accuracy in our comparison (Table 5). Six of the cured patients had recurrent hyperparathyroidism. Median time from their initial surgery to recurrence was 3 years.
DISCUSSION MIP guided by IOPTH performed in patients with sporadic hyperparathyroidism had an operative success rate of 96.6 % with less than 10 % BCE. Ten additional patients achieved operative success after a second operation, for an overall success rate of 98.4 % (546 of 557). The most common cause of surgical failure (11 of 19) in our study was missed MGD by operating surgeons. Falsely positive IOPTH results were responsible for the remaining 8 surgical failures.
S. Lee et al. TABLE 2 Operative details of 19 patients who failed the initial MIP and outcome of the second operation Patient no. Reason for failure
IOPTH Initial decrease surgery (%)
Quadrant Second from operation/reason which glands for no surgery removed
1
Fail to identify gland
0
MIP to BCE LS
BCE
None
Unable to identify S
2
Fail to identify gland
13
MIP to BCE LS
BCE
RI
Multigland disease
S
3
Fail to identify gland
30
MIP to BCE LS, RS, LI
BCE
LI
Multigland disease
S
4
Surgeon decision 3
MIP
LS, LI
MIP
RS, 1/2RI
Multigland disease
S
5
Surgeon decision 25
MIP
LS
BCE
RI, RS
Multigland disease
S
6
Surgeon decision 39
MIP
RS
BCE
LS, LI
Multigland disease
S
7
Surgeon decision 3
MIP
RI
BCE
LI, 1/2LS
Multigland disease
S
8
Surgeon decision 0
MIP
LI
MIP
LI
Incomplete excision
S
9
False IOPTH
53
MIP
LS
BCE
LI
Multigland disease
S
10
False IOPTH
61
MIP
LI
MIP
RI
Multigland disease
S
11
False IOPTH
57
MIP
RS, RI
MIP
LI, 1/2LS
Multigland disease
F
12
Fail to Identify gland
0
MIP
Medical management
13
Surgeon decision 0
MIP
Medical reason
14
Surgeon decision 23
MIP
Surgery planned
15
False IOPTH
55
MIP
Medical reason
16
False IOPTH
56
MIP
Medical reason
17
False IOPTH
64
MIP
No longer met criteria
18
False IOPTH
50
MIP
No longer met criteria
19
False IOPTH
58
MIP
Patient did not follow up
Quadrant from Operative finding which glands removed, second operation
Outcome
IOPTH intraoperative parathyroid hormone, MIP minimally invasive parathyroidectomy, BCE bilateral cervical exploration, LS left superior parathyroid, LI left inferior parathyroid, RS right superior parathyroid, RI right inferior parathyroid, S operative success, F operative failure
Of the 11 operative failures that were the result of missed MGD by surgeons, 7 cases were due to surgeon’s operative decision making by terminating the operation with an inadequate decline in IOPTH and not performing BCE. Clinical decision making for early termination for these 7 patients included removal of radiographically concordant suspected single-gland disease with identification of a normal ipsilateral second parathyroid gland. Other factors (such as a desire to minimize anesthetic time) also led to termination of the procedure. Because the IOPTH assays are performed in the laboratory, which is remote from the operating theater, there is a delay. Of these 7 patients, 5 underwent second operations. Four patients had unrecognized MGD in the initial operation, and 1 patient
had incomplete excision of a dumbbell-shaped, single abnormal parathyroid gland. The remaining 4 operative failures were the result of surgeons’ combined skill and judgment. These 4 patients underwent BCE, and surgeon failed to identify all abnormal parathyroid glands. Lew et al. reviewed a total of 723 patients with 21 operative failures (2.9 %). The main cause of their operative failures was an inability to find the abnormal parathyroid glands (16 of 21 patients, 76 %).11 Our study also confirms their finding that surgeons’ clinical skills and judgment play an important role in the surgical outcome of MIP. The remainder of the operative failures (8 of 19) were the result of falsely positive IOPTH results. IOPTH decreased 50 % or more without operative success in these
Operative Failure in Minimally Invasive Parathyroidectomy
cases. In the present study, we found that the percentage decrease was important because patients with robust PTH declines were more likely to achieve operative success. We found that 80 % of patients will have greater than 70 % IOPTH decrease with 0 % failure rate, whereas a decrease of 50–59 % had a failure rate of 20 %. Adding the additional criterion of PTH levels falling into the normal range in addition to IOPTH decrease of 50 % or more did not seem to improve surgical outcome significantly in our study (Tables 4, 5). Increasing the IOPTH decrease level to 60 % or more also did not improve the accuracy of the test (Table 5). Previous studies have shown that stricter criteria TABLE 3 IOPTH decrease and operative outcome Level of IOPTH decrease (%) Operative success Operative failure (n = 528) (n = 19) \50
15 (57.7 %)
11 (42.3 %)
50–59
24 (80.0 %)
6 (20.0 %)
60–69
67 (97.1 %)
2 (2.9 %)
C70
422 (100.0 %)
0 (0.0 %)
IOPTH intraoperative parathyroid hormone
TABLE 4 Comparison of IOPTH criteria IOPTH status
Operative success n = 538 (96.6 %)
Operative failure n = 19 (3.4 %)
Yes
513 (97.2)
8 (42.1)
No
15 (2.8)
11 (57.9)
P value
\0.001
IOPTH decrease C50 %
\0.001
IOPTH decrease C60 % Yes
489 (92.6)
2 (10.5)
No
39 (7.4)
17 (89.5) \0.001
IOPTH decrease C50 % and PTH B80 Yes
482 (91.3)
5 (26.3)
No
46 (8.7)
14 (73.7)
IOPTH intraoperative parathyroid hormone
do not significantly improve the operation’s success rate.12 However, with a failure rate close to 20 % in the group with a lower IOPTH decrease (50–59 %), selective BCE can be considered in patients with marginal IOPTH decrease. Among the cohort of patients who underwent a BCE, the decision to convert to bilateral surgery was prompted by IOPTH in 67 % (31 of 46) of the patients. In all of these cases, unsuspected additional abnormal parathyroid glands were identified. Other patients underwent BCE on the basis of the surgeon’s clinical suspicion of MGD, such as normal parathyroid gland by surgeon visual inspection on the explored side. Previous studies showed that surgical experience plays a critical role in the overall success of parathyroid surgery.13 We believe that the surgeon’s experience, augmented by IOPTH, can decrease the need for unselective BCE while still increasing the identification of abnormal parathyroid gland. Routine BCE would not improve the successful outcome significantly because the false-positive rate of IOPTH is very low (1.4 %, 8 of 557). Even though BCE can be performed with minimal complications in experienced hands, complications such as recurrent laryngeal nerve injury and hypoparathyroidism do occur. Patients who had MIP only put one side of the neck structure at risk. The potential benefit of performing MIP is that the contralateral dissection can be performed in the future in a region without scarring or an increased risk of nerve injury, unlike a procedure being performed after initial BCE would. Patients in our study who had operative failure had higher mean preoperative serum calcium and creatinine levels. However, it is difficult to judge the clinical significance of these findings. An element of secondary hyperparathyroidism and/or higher prevalence of parathyroid hyperplasia or MGD may be responsible. Age, sex, race, body mass index, and preoperative PTH levels did not differ significantly between the patients who had successful surgeries and those whose surgeries failed. This study has several limitations. As a retrospective descriptive study, hypotheses that test comparisons of MIP to BCE cannot be performed. Fortunately, operative failure
TABLE 5 IOPTH accuracy rates with different IOPTH criteria IOPTH decrease
Sensitivity (TP/TP ? FN)
Specificity (TN/TN ? FP)
PPV (TP/TP ? FP)
NPV (TN/TN ? FN)
Accuracy (TP ? TN/ TP ? TN ? FP ? FN)
C50 %
97.2
57.9
98.5
42.3
95.8
C60 %
92.6
89.5
99.6
30.4
92.5
C50 % and PTH B80
91.3
73.7
99.0
23.3
90.7
IOPTH intraoperative parathyroid hormone, TP truly positive—correct prediction of postoperative cure, TN truly negative—correct prediction of postoperative failure, FP falsely positive—incorrect prediction of postoperative cure, FN falsely negative—incorrect prediction of postoperative failure, PPV positive predictive value, NPV negative predictive value
S. Lee et al.
after parathyroid surgery is not common. To better understand the cause of failure, a large number of patients who undergo parathyroidectomy is needed. Therefore, even though this study is descriptive in nature, it can contribute the better understanding of the reason for failure in MIP. Second, our follow-up data are suboptimal. More than 30 % of our patients did not complete 6-month follow-up, likely as a result of a geographically diverse referral base. All patients who did not complete 6-month follow-up, however, had documented normal calcium level postoperatively. In conclusion, a targeted MIP guided by preoperative radiological localization and IOPTH can achieve an excellent operative success rate without the need for routine BCE. The most common cause of operative failure of MIP in our study was surgeons’ inability to identify all abnormal parathyroid glands. Patients who have more than 70 % IOPTH decrease are likely cured and do not need BCE. When there is an IOPTH decrease of more than 50 % but less than 60 %, the surgeon should use his or her judgment as to whether the patient needs BCE. ACKNOWLEDGMENT Supported in part by The MD Anderson Cancer Center Support Grant CA016672. We thank Zach Bohannan and Melissa Burkett from The University of Texas MD Anderson Cancer Center Department of Scientific Publications for their assistance. DISCLOSURE
The authors declare no conflict of interest.
REFERENCES 1. Piovesan A, Molineri N, Casasso F, et al. Left ventricular hypertrophy in primary hyperparathyroidism. Effects of successful parathyroidectomy. Clin Endocrinol (Oxf). 1999;50:321–8.
2. Rubin MR, Bilezikian JP, McMahon DJ, et al. The natural history of primary hyperparathyroidism with or without parathyroid surgery after 15 years. J Clin Endocrinol Metab. 2008;93:3462– 70. 3. Edwards ME, Rotramel A, Beyer T, et al. Improvement in the health-related quality-of-life symptoms of hyperparathyroidism is durable on long-term follow-up. Surgery. 2006;140:655–63. 4. Espiritu RP, Kearns AE, Vickers KS, Grant C, Ryu E, Wermers RA. Depression in primary hyperparathyroidism: prevalence and benefit of surgery. J Clin Endocrinol Metab. 2011;96:E1737–45. 5. Irvin GL 3rd, Carneiro DM. Management changes in primary hyperparathyroidism. JAMA. 2000;284:934–6. 6. Irvin GL 3rd, Carneiro DM, Solorzano CC. Progress in the operative management of sporadic primary hyperparathyroidism over 34 years. Ann Surg. 2004;239:704–8. 7. Udelsman R, Lin Z, Donovan P. The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism. Ann Surg. 2011;253:585–91. 8. Greene AB, Butler RS, McIntyre S, et al. National trends in parathyroid surgery from 1998 to 2008: a decade of change. J Am Coll Surg. 2009;209:332–43. 9. Norman J, Lopez J, Politz D. Abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid operations. J Am Coll Surg. 2012;214:260–9. 10. Siperstein A, Berber E, Barbosa GF, et al. Predicting the success of limited exploration for primary hyperparathyroidism using ultrasound, sestamibi, and intraoperative parathyroid hormone: analysis of 1158 cases. Ann Surg. 2008;248:420–8. 11. Lew JI, Rivera M, Irvin GL 3rd, Solorzano CC. Operative failure in the era of focused parathyroidectomy: a contemporary series of 845 patients. Arch Surg. 2010;145:628–33. 12. Barczynski M, Konturek A, Hubalewska-Dydejczyk A, Cichon S, Nowak W. Evaluation of Halle, Miami, Rome, and Vienna intraoperative iPTH assay criteria in guiding minimally invasive parathyroidectomy. Langenbecks Arch Surg. 2009;394:843–9. 13. Chen H, Wang TS, Yen TW, et al. Operative failures after parathyroidectomy for hyperparathyroidism: the influence of surgical volume. Ann Surg. 2010;252:691–5.
