Eur Arch Otorhinolaryngol DOI 10.1007/s00405-015-3678-6
HEAD AND NECK
Risk factors for treatment failure in surgery for primary hyperparathyroidism: the impact of change in surgical strategy and training procedures Anders Rørbæk Madsen1 • Lars Rasmussen2 • Christian Godballe1
Received: 26 February 2015 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Surgery for primary hyperparathyroidism (pHPT) has a high cure-rate and few complications. Preoperative localization procedures have permitted a dramatic shift from routine bilateral exploration to focused, minimally invasive procedures. At Odense University Hospital, Denmark, the introduction of focused surgery was combined with training of new surgeons. The objective of this study was to identify possible risk factors for treatment failure with special focus on surgical strategy and training of new surgeons. A 6-year prospective and consecutive series of 567 pHPT patients operated at Odense University hospital, Denmark, was analyzed. A shift in strategy was made in 2006 and at the same time new surgeons started training in parathyroid surgery. Biochemical-, clinical- and follow-up data were analyzed. Overall cure-rate was 90.7 %. Complication rates were 1.1 % for hemorrhage, 1.1 % for wound infection and 0.9 % for recurrent nerve paralysis. The only significant predictor of treatment failure at 6 months was histology of hyperplasia (OR 4.3). Neither the introduction of minimal invasive surgical strategy nor the training of new surgeons had a significant influence on the rate of treatment failures. Hyperplasia is a significant predictor of treatment failure in pHPT surgery. A shift towards systematic preoperative localization with focused surgery as well as training of new surgeons can be done without negative impact on treatment results. Identification of the hyperplasia and multigland patients in need of
& Anders Rørbæk Madsen [email protected] 1
Department of ENT Head and Neck Surgery (F), Odense University Hospital, Sdr. Boulevard, 5000 Odense, Denmark
2
Department of Surgery, Odense University Hospital, Odense, Denmark
bilateral cervical exploration is crucial to avoid failures and raise cure rates. Keywords Primary hyperparathyroidism Focused surgery Hypercalcaemia Preoperative localization Training Strategy
Introduction Hyperparathyroidism is caused by excessive hormone secretion from the parathyroid glands from either intrinsic (primary, tertiary) or extrinsic (secondary) reasons. The majority of cases presented to surgical departments are primary hyperparathyroidism (pHPT). While often asymptomatic and discovered by routine blood screening, some cases are still diagnosed due to symptoms such as osteoporosis, kidney or urinary tract stones, muscle symptoms or fatigue [1]. Furthermore, it has recently been shown that even mild hypercalcaemia can impair cognitive functions, and surgery should also be considered for this group [2]. Parathyroid surgery performed on the right indication has a long term beneficial influence on quality of life [3]. Selected cases with only mild hypercalcaemia can, as an alternative, be observed or treated medically, for example, with calcimimetics. However, the only curative treatment is surgical removal of the autonomic parathyroid tissue. In general, surgery of pHPT has a high cure-rate and few complications [4]. To continue or even improve the good results and to secure a sufficient number of surgeons, it is necessary to optimize surgical strategies and train new surgeons. This might, however, influence outcome. We present a prospective and consecutive series of patients treated for pHPT at Odense University Hospital in
123
Eur Arch Otorhinolaryngol
Denmark from 2004 to 2009. In 2006, a shift from bilateral cervical exploration (BCE) to focused, minimally invasive procedures was made and at the same time surgeons from the Department of ENT Head and Neck Surgery (ENT HNS) started training in parathyroid surgery. Until then, the operations had been performed by only one experienced general surgeon. These changes open a unique possibility to analyze the impact from a shift in surgical strategy and introduction of a learning program on the risk of treatment failure. The objective of this study was to identify possible risk factors for treatment failure with special focus on surgical strategy and training of new surgeons.
Materials and methods During the 6 year period from January 2004 to December 2009, preoperative, perioperative and postoperative data were registered prospectively and consecutively in online database (Scandinavian Quality register for Thyroid and Parathyroid Surgery) [5]. A total of 567 patients were surgically treated for pHPT—445 women (78 %) and 122 men (22 %). Median age was 60 years (range 19–91). Symptomatic disease was found in 58 % of patients and preoperative ionized S-calcium levels ranged from 1.31 to 2.27 (median 1.49, normal reference range 1.18–1.32). Sporadic disease was predominating with 550 patients (97 %) while 13 patients (2 %) had MEN1 and 2 patients hereditary HPT (0.4 %), and 2 patients (0.4 %) hypophosphatic rachitis. No patients had MEN2A. Familial benign hypocalciuric hypercalcemia (FHH) was excluded preoperatively on all patients by urine analysis (calcium/creatinine ratio) or by testing of the CaSR gene by blood samples. History of preoperative lithium treatment was not recorded. Ten patients (0.2 %) had thyroid surgery performed earlier. Laryngoscopy was made pre- and post-operatively on a routine basis (99.3 and 99.1 %, respectively). All operations were performed by consultant surgeons. Fifty patients (9 %) had their surgery performed in a learning setup with an experienced general surgeon as teache, and a head and neck surgeon as learner. The method of teaching the ENT surgeons was traditional surgical supervision (apprenticeship) with the ENT surgeon as primary surgeon and the experienced general surgeon as teacher/supervisor. In 477 (84 %) of all cases, preoperative localizations techniques were used—ultrasound in 457 patients (81 %), Sestamibi 399 (70 %), CT-scan 3 (0.5 %), PET 3 (0.5 %) and MRI 2 (0.4 %).
123
All surgery was done under general anesthesia and as inhospital procedures. Frozen-section histology was used in 505 (89 %) of cases. SPECT-CT and intraoperative parathyroid hormone measurements (IOPTH) were not used during this period. Follow-up was made with clinical examination including routine laryngoscopy, as well as measurement of calcium levels after 6 weeks and 6 months. At the 6 months follow-up, calcium status was recorded in 549 patients (96.8 %). Thyroid surgery (unilateral resection or lobectomy) was performed during parathyroid surgery in 63 patients (11 %), whereas thymus was partly or completely resected in 30 cases (5 %). Sternotomy was not necessary in any cases. Data were statistically analyzed in the medical database and analysis program ‘‘Medlog 2008-2’’ and in Microsoft Excel. v2-distribution test was used. Odds ratio (OR) was interpreted as the relative risk (RR).
Results BCE was performed in 287 cases (51 %), focused operation (visualization of only one gland) in 192 patients (34 %) and unilateral exploration (visualization of two glands unilaterally) in 88 (16 %) of cases. In 263 (92 %) of the BCEs, the procedure was planned preoperatively while 24 (8 %) were converted from focused or unilateral operations. In 498 patients (90.7 %), normocalcaemia was registered at follow-up blood testing, while 51 (9.3 %) had spontaneous hypercalcaemia interpreted as persistent disease. Complications Reoperation for postoperative hemorrhage was necessary in 6 patients (1.1 %) and postoperative wound infection was found in 6 patients (1.1 %). Preoperative laryngoscopy was performed in 563 patients (99.3 %). Combining follow-up data from the visits after 6 weeks and 6 months postoperative laryngoscopy was registered in 562 patients (99.1 %). In 6 patients (1.1 %), unilateral recurrent nerve paralysis caused by the operation was found. In 5 cases, no laryngoscopy was performed. No cases of bilateral paresis were found. As 287 procedures were bilateral, a total of 854 nerves were at risk—resulting in 0.7 % recurrent nerve paralysis of nerves at risk. Table 1 shows the complication rates for the overall series and for different types of surgery. While there is a tendency towards more frequent postoperative
Eur Arch Otorhinolaryngol Table 1 Complication rates for focused, unilateral and bilateral operations
Focused
Patients
Hemorrhage
Post infection
Recurrent nerve paresis 1 (0.5 %) NS
192
2 (1.0 %) NS
1 (0.5 %) NS
Unilateral
88
1 (1.1 %) NS
0 (0 %) NS
0 (0 %) NS
Bilateral
287
3 (1.1 %) NS
5 (1.7 %) NS
5 (1.7 %) (0.9 %NAR) NS
Overall
567
6 (1.1 %)
6 (1.1 %)
6 (1.1 %) (0.7 %NAR) NS
RNP recurrent nerve paresis, NAR nerves at risk, NS not significant
infections and recurrent nerve paresis in bilateral surgery, there is no significant difference. Histology In 515 patients (91 %), only one parathyroid gland was removed. Figure 1 shows the distribution of excised parathyroid glands. The median weight of parathyroid tissue removed was 0.43 g (range 0.05–7.69 g). Frozen sections were correct in 499 cases (99 %) compared to final histology. Parathyroid adenoma unfortunately lacks a precise histopathological definition [6, 7]. In this study, the diagnosis of adenoma was based on the presence of adjacent rim of normal parathyroid cells in the resected gland or a per-operative macroscopic or biopsy verified identification of normal or suppressed parathyroid glands. Final histology showed parathyroid adenoma in 435 cases (77 %), hyperplasia in 100 cases (18 %) and 2 lipo-adenomas (0.4 %). In 30 (6 %) of the explorations, no parathyroid tissue was found. These cases were all BCEs and were not excluded from analysis. Follow-up of these 30 patients interestingly showed persistent disease/failure in 52 % (18/34) of cases. No parathyroid cancers were diagnosed. Preoperative localization During the beginning of the study period, preoperative diagnostic localization procedures were not performed on a routine basis. From April 2006, preoperative 493
Paents
30
16 0 glands
1
2
7 3
18 Other
Fig. 1 Number of excised glands—493 (87 %) had one parathyroid gland removed. In 30 cases final histology showed no parathyroid tissue removed
ultrasonography as well as 99mTc Sestamibi scintigraphy (Sestamibi) was implemented as routine for all pHPT patients. To analyze the value of these measures we divided the material into two subgroups—before and after the 1st of April 2006. As demonstrated in Table 2, the implementation of routine preoperative localization procedures increased the rate of focused procedures by ten times and the rate of unilateral operations by 2.5 times. The cure-rate for the period 2004–2006 was 94.4 % and from 2006 to 2010 it was 88.7 %. The difference was not significant. Overall, during the 6-year period, the Sestamibi localized single gland disease correctly in 75 % of patients and ultrasound correctly in 61 % as described in Table 3. In 214 (37 %) cases, both ultrasonography and Sestamibi were correct and in this group, 97 % were cured by primary surgery. Risk factors for failure Table 4 shows univariate analysis of selected variables concerning risk of failure defined as hypercalcaemia 6 months postoperatively. Histology of hyperplasia was statistically significant with an odds ratio of 4.3 and p value \0.001. No other significant predictors of failure were found. A strong tendency towards failure was found in a small number of 29 patients with previous parathyroid surgery. In 6/29 (20.7 %) of cases, failure was found resulting in an OR of 2.5 (ns) (Table 4). In an attempt to preoperatively identify cases of hyperplasia—the patients with high risk of failure—further analysis was made. Table 5 shows analysis of factors related to hyperplasia. However, we did not find any significant preoperative factors to identify patients with hyperplasia. Learning setup The 50 patients who had surgery performed in a learning setup including experienced general surgeon as teacher and a younger surgeon as learner were identified retrospectively. None of these had recurrent nerve palsy or
123
Eur Arch Otorhinolaryngol Table 2 A remarkable increase of focused and unilateral operations followed the implementation of routine preoperative ultrasound and Sestamibi. Overall cure-rate was 90.7 %
Jan 2004–Mar 2006
Apr 2006–Dec 2009
Patients
197
370
Ultrasound
49 % (97)
99.5 % (368)
SM-scint.
19 % (38)
98 % (363)
US ? SM
8 % (15)
98 % (361)
Bilat expl
88 % (173)
31 % (114)
Unilat expl
8 % (15)
20 % (73)
Focused
5 % (9)
50 % (183)
Cure-rate (NS)
94.3 % (183/194) 3 unknown
88.7 % (314/354) 16 unknown
Reoperations
5 % (10)
5 % (19)
SM Sestamibi scintigraphy, US ultrasound, NS not significant
Table 3 Accuracy of preoperative ultrasound and Sestamibi scintigraphy localization diagnosis Localization
Sestamibi
Ultrasound
Correct Partially correct (multigland disease)
77 % (307) 0.5 % (2)
61 % (279) 1 % (5)
Inconclusive
17 % (68)
33 % (151)
6 % (22)
5 % (22)
False
In 214 (37 %) cases, both ultrasonography and Sestamibi were correct—in this group, 97 % were cured by primary surgery
postoperative infection. One patient (2 %) had a postoperative hemorrhage in need of acute reoperation. The lower complication rate in the learning setup patients was not significant. In three of the 50 ‘‘learningpatients’’, hypercalcaemia was found 6 months after surgery—resulting in a 94 % cure-rate.
Discussion Our introduction of Sestamibi and ultrasonography as standard preoperative localization examinations increased the use of unilateral explorations by 2.5 times and the use of focused surgery by 10 times. Though the duration of surgery was not measured in our study, the subsequent change towards unilateral and focused surgery has clearly decreased the duration of the operations and has allowed us to operate more patients per day. Cure rate and complications The 567 patients with primary hyperthyroidism were offered an effective and low risk treatment—overall, surgery cured 90.7 % of cases with low complication rates (Table 1). The change from routine BCE exploration to unilateral and focused surgery was, however, followed by
123
an unsatisfactory fall in cure-rate from 94.3 % before 2006 to 88.7 % from 2006. Though not significant, the fall in cure-rate after our introduction of focused surgery indicates that an increased focus on potential hyperplasia cases has to be introduced and that the threshold for conversion to BCE has to be low. A tendency towards lower complications rates was observed in unilateral and focused procedures but no significant difference was found, presumably due to a small number of patients (type II error). Focused and unilateral surgery is generally accepted as having lower morbidity and is widely considered to be the gold standard of parathyroid surgery today [4, 8]. Our results, however, show that hyperplasia (multigland disease) is a significant predictor of failure to cure the patients and that minimally invasive focused surgery may lead to increased risk of failure due to missed cases of hyperplasia. The high failure rate in patients where final histology showed no parathyroid tissue removed underlines the importance of correct preoperative localization and meticulous dissection—and emphasizes the potential of intraoperative PTH measurement to let the surgeon know when to stop the surgery. Literature In order to evaluate our results we selected comparable recent studies—an overview hereof is found in Table 6. Very high cure rates have been published from single institutions. Routine IOPTH seems to be a valuable tool to intraoperatively identify hyperplasia/multigland patients in need of BCE and thereby to increase cure rates. This has also been reported earlier by Gill et al. [9]. Bergenfelz et al. [5] have published a large multicenter Scandinavian study of 2708 patients to which we have also contributed. IOPTH was used in 29 % of cases and was found to significantly increase the chance of cure. In 2007, Allendorf et al. [10] published a single institution, one surgeon 17-year material of 1112 patients. All patients had BCE—in fact, 76 % of the procedures were performed in local anesthesia.
Eur Arch Otorhinolaryngol Table 4 Univariate analysis of selected variables concerning risk of failure Variable Age Gender Previous parathyroid surgery Type of exploration
Dichotomization
Observations
Failures
Percentage
p value
Odds ratio
NS
1.2
NS
1.7
NS
2.5
NS
1.10
NS
2.7
\0.001
4.3
NS
0.8
NS
1.4
NS
0.6
NS
1.0
NS
0.9
NS
1.1
Percentage
p value
Odds ratio
NS
0.8
NS
0.7
NS
1.0
NS
1.5
NS
1.0
NS
0.7
C60 years
282
28
9.9
\60 years
284
23
8.1
Male
121
16
13.2
Female
445
35
7.9
Yes
29
6
20.7
No
537
45
8.4
Unilateral
279
27
9.4
Bilateral
287
24
8.6
4
23.5
Hereditary endocrine disease
Yes
17
No
549
47
8.6
Histology
Hyperplasia Adenoma
99 434
16 16
16.2 3.7
Teaching Setup
Yes
52
4
7.7
No
514
47
9.1
Yes
476
45
9.5
No
90
6
6.7
Yes
329
23
7.0
No
237
28
11.8
Yes
274
25
9.1
No
292
26
8.9
Yes
138
11
8.0
No
428
40
9.4
Localization examinations preop. Symptomatic disease Ca?? level preoperatively above median (1.49) Ca?? level preoperatively [1.56 (upper quartile) Ca?? level preoperatively \1.44 (lower quartile)
Yes
157
15
9.6
No
409
36
8.8
Failure was defined as hypercalcaemia 6 months postoperatively
Table 5 Analysis of factors related to hyperplasia Variable Age Gender Symptomatic disease Preoperative calcium above median (1.49) Ca?? level preoperatively [1.56 (upper quartile) Ca?? level preoperatively \1.44 (lower quartile)
Dichotomization
Observations
Hyperplasia
C60 years
282
44
15.6
\60 years
285
56
19.7
Male
445
71
16.0
Female
122
29
23.8
Yes
330
58
17.6
No
237
42
17.7
Yes
274
59
21.5
No
293
41
14.0
Yes
138
25
18.1
No
429
75
17.5
Yes No
157 410
20 80
12.7 19.5
No significant factors were found
Siperstein et al. [11] found that 16 % of cases of multigland disease were overseen due to limitations of localization techniques. Lew et al. [12] analyzed this further and
concluded that inability of the surgeon rather than multigland disease was the reason for failure—at least in the presence of IOPTH. In a large study of 15,000 parathyroid
123
Eur Arch Otorhinolaryngol Table 6 Overview of comparable current studies of surgery for pHPT Author
Patients/period
Cure-rate (%)
Complication rates
Remarks
Bergenfelz et al. [5]
2708 (2004–2008)
93
0.8 % postop. bleeding, 1.2 % infection and 1.4 % unilateral recurrent nerve palsy
Multicenter, Scandinavia. IoPTH in 29 % of cases
Allendorf et al. [10]
1112 (1987–2003)
97.4
0.2 % recurrent nerve injury and postop. bleeding of 0.8 %
Single institution, New York. Single surgeon. 76 % in local anesthesia
Siperstein et al. [11]
1158 (1999–2007)
98
Not described
Single institution, Ohio. All bilateral explorations, IOPTH in use
Lew et al. [12]
845 (1993–2009)
97.1
Not described
Single institution, Miami. All IOPTH
Lee et al. [14]
538 (1998–2010)
96.6
Not described
Single institution, Texas, All IOPTH
Present study
567 (2004–2009)
90.7
1.1 % for bleeding, 1.1 % for wound infection and 0.9 % for recurrent nerve paralysis
Single institution, Denmark. IOPTH not in use. 99 % laryngoscopy pre and postop.
operations, Norman et al. [13] had similar findings and argued for a return to routine BCE to reduce failures. In a recent review of 538 patients, Lee et al. [14] found a cure-rate of 96.6 % with focused surgery. In a systematic analysis of their failures, they found multigland disease, falsely positive IOPTH levels, and surgeon’s inability to identify the abnormal parathyroid glands to be the causes of their failures. Their advice was to consider conversion to BCE in selected cases with low IOPTH level drops.
Conclusions No increase in complication rates was detected during the time period in which ENT head and neck surgeons were trained in parathyroid surgery, indicating that a shift between specialties is possible without any negative impact on surgical quality. Routine presurgical diagnostics makes focused surgery feasible but the price might be an increased failure rate—primarily because of not diagnosed parathyroid hyperplasia. Conflict of interest of interest.
The authors declare that they have no conflict
Ethical standard All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from all individual participants included in the study.
References 1. Fraser WD (2009) Hyperparathyroidism. Lancet 374(9684): 145–158. doi:10.1016/S0140-6736(09)60507-9
123
2. Babinska D, Barczynski M, Stefaniak T, Oseka T, Babinska A, Babinski D, Sworczak K, Lachinski AJ, Nowak W, Sledzinski Z (2012) Evaluation of selected cognitive functions before and after surgery for primary hyperparathyroidism. Langenbecks Arch Surg 397(5):825–831. doi:10.1007/s00423-011-0885-5 3. Edwards ME, Rotramel A, Beyer T, Gaffud MJ, Djuricin G, Loviscek K, Solorzano CC, Prinz RA (2006) Improvement in the health-related quality-of-life symptoms of hyperparathyroidism is durable on long-term follow-up. Surgery 140(4):655–663. doi:10. 1016/j.surg.2006.06.016 (discussion 653–654) 4. Udelsman R, Lin Z, Donovan P (2011) The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism. Ann Surg 253(3):585–591. doi:10.1097/SLA.0b013e318208fed9 5. Bergenfelz AO, Jansson SK, Wallin GK, Martensson HG, Rasmussen L, Eriksson HL, Reihner EI (2009) Impact of modern techniques on short-term outcome after surgery for primary hyperparathyroidism: a multicenter study comprising 2,708 patients. Langenbecks Arch Surg 394(5):851–860. doi:10.1007/ s00423-009-0540-6 6. Ghandur-Mnaymneh L, Kimura N (1984) The parathyroid adenoma. A histopathologic definition with a study of 172 cases of primary hyperparathyroidism. Am J Pathol 115(1):70–83 7. Erickson LA, Jin L, Wollan P, Thompson GB, van Heerden JA, Lloyd RV (1999) Parathyroid hyperplasia, adenomas, and carcinomas: differential expression of p27Kip1 protein. Am J Surg Pathol 23(3):288–295 8. Kunstman JW, Udelsman R (2012) Superiority of minimally invasive parathyroidectomy. Adv Surg 46:171–189 9. Gill MT, Dean M, Karr J, Aultman DF, Nathan CO (2011) Intraoperative parathyroid hormone assay: a necessary tool for multiglandular disease. Otolaryngol Head Neck Surg 144(5):691–697. doi:10.1177/0194599811398597 10. Allendorf J, DiGorgi M, Spanknebel K, Inabnet W, Chabot J, Logerfo P (2007) 1112 consecutive bilateral neck explorations for primary hyperparathyroidism. World J Surg 31(11):2075–2080. doi:10.1007/s00268-007-9068-5 11. Siperstein A, Berber E, Barbosa GF, Tsinberg M, Greene AB, Mitchell J, Milas M (2008) Predicting the success of limited exploration for primary hyperparathyroidism using ultrasound, sestamibi, and intraoperative parathyroid hormone: analysis of 1158 cases. Ann Surg 248(3):420–428. doi:10.1097/SLA. 0b013e3181859f71 12. Lew JI, Rivera M, Irvin GL 3rd, Solorzano CC (2010) Operative failure in the era of focused parathyroidectomy: a contemporary
Eur Arch Otorhinolaryngol series of 845 patients. Arch Surg 145(7):628–633. doi:10.1001/ archsurg.2010.104 13. Norman J, Lopez J, Politz D (2012) Abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid operations. J Am Coll Surg 214(3):260–269. doi:10. 1016/j.jamcollsurg.2011.12.007
14. Lee S, Ryu H, Morris LF, Grubbs EG, Lee JE, Harun N, Feng L, Perrier ND (2014) Operative failure in minimally invasive parathyroidectomy utilizing an intraoperative parathyroid hormone assay. Ann Surg Oncol. doi:10.1245/s10434-013-3479-3
123
HEAD AND NECK
Risk factors for treatment failure in surgery for primary hyperparathyroidism: the impact of change in surgical strategy and training procedures Anders Rørbæk Madsen1 • Lars Rasmussen2 • Christian Godballe1
Received: 26 February 2015 / Accepted: 27 May 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Surgery for primary hyperparathyroidism (pHPT) has a high cure-rate and few complications. Preoperative localization procedures have permitted a dramatic shift from routine bilateral exploration to focused, minimally invasive procedures. At Odense University Hospital, Denmark, the introduction of focused surgery was combined with training of new surgeons. The objective of this study was to identify possible risk factors for treatment failure with special focus on surgical strategy and training of new surgeons. A 6-year prospective and consecutive series of 567 pHPT patients operated at Odense University hospital, Denmark, was analyzed. A shift in strategy was made in 2006 and at the same time new surgeons started training in parathyroid surgery. Biochemical-, clinical- and follow-up data were analyzed. Overall cure-rate was 90.7 %. Complication rates were 1.1 % for hemorrhage, 1.1 % for wound infection and 0.9 % for recurrent nerve paralysis. The only significant predictor of treatment failure at 6 months was histology of hyperplasia (OR 4.3). Neither the introduction of minimal invasive surgical strategy nor the training of new surgeons had a significant influence on the rate of treatment failures. Hyperplasia is a significant predictor of treatment failure in pHPT surgery. A shift towards systematic preoperative localization with focused surgery as well as training of new surgeons can be done without negative impact on treatment results. Identification of the hyperplasia and multigland patients in need of
& Anders Rørbæk Madsen [email protected] 1
Department of ENT Head and Neck Surgery (F), Odense University Hospital, Sdr. Boulevard, 5000 Odense, Denmark
2
Department of Surgery, Odense University Hospital, Odense, Denmark
bilateral cervical exploration is crucial to avoid failures and raise cure rates. Keywords Primary hyperparathyroidism Focused surgery Hypercalcaemia Preoperative localization Training Strategy
Introduction Hyperparathyroidism is caused by excessive hormone secretion from the parathyroid glands from either intrinsic (primary, tertiary) or extrinsic (secondary) reasons. The majority of cases presented to surgical departments are primary hyperparathyroidism (pHPT). While often asymptomatic and discovered by routine blood screening, some cases are still diagnosed due to symptoms such as osteoporosis, kidney or urinary tract stones, muscle symptoms or fatigue [1]. Furthermore, it has recently been shown that even mild hypercalcaemia can impair cognitive functions, and surgery should also be considered for this group [2]. Parathyroid surgery performed on the right indication has a long term beneficial influence on quality of life [3]. Selected cases with only mild hypercalcaemia can, as an alternative, be observed or treated medically, for example, with calcimimetics. However, the only curative treatment is surgical removal of the autonomic parathyroid tissue. In general, surgery of pHPT has a high cure-rate and few complications [4]. To continue or even improve the good results and to secure a sufficient number of surgeons, it is necessary to optimize surgical strategies and train new surgeons. This might, however, influence outcome. We present a prospective and consecutive series of patients treated for pHPT at Odense University Hospital in
123
Eur Arch Otorhinolaryngol
Denmark from 2004 to 2009. In 2006, a shift from bilateral cervical exploration (BCE) to focused, minimally invasive procedures was made and at the same time surgeons from the Department of ENT Head and Neck Surgery (ENT HNS) started training in parathyroid surgery. Until then, the operations had been performed by only one experienced general surgeon. These changes open a unique possibility to analyze the impact from a shift in surgical strategy and introduction of a learning program on the risk of treatment failure. The objective of this study was to identify possible risk factors for treatment failure with special focus on surgical strategy and training of new surgeons.
Materials and methods During the 6 year period from January 2004 to December 2009, preoperative, perioperative and postoperative data were registered prospectively and consecutively in online database (Scandinavian Quality register for Thyroid and Parathyroid Surgery) [5]. A total of 567 patients were surgically treated for pHPT—445 women (78 %) and 122 men (22 %). Median age was 60 years (range 19–91). Symptomatic disease was found in 58 % of patients and preoperative ionized S-calcium levels ranged from 1.31 to 2.27 (median 1.49, normal reference range 1.18–1.32). Sporadic disease was predominating with 550 patients (97 %) while 13 patients (2 %) had MEN1 and 2 patients hereditary HPT (0.4 %), and 2 patients (0.4 %) hypophosphatic rachitis. No patients had MEN2A. Familial benign hypocalciuric hypercalcemia (FHH) was excluded preoperatively on all patients by urine analysis (calcium/creatinine ratio) or by testing of the CaSR gene by blood samples. History of preoperative lithium treatment was not recorded. Ten patients (0.2 %) had thyroid surgery performed earlier. Laryngoscopy was made pre- and post-operatively on a routine basis (99.3 and 99.1 %, respectively). All operations were performed by consultant surgeons. Fifty patients (9 %) had their surgery performed in a learning setup with an experienced general surgeon as teache, and a head and neck surgeon as learner. The method of teaching the ENT surgeons was traditional surgical supervision (apprenticeship) with the ENT surgeon as primary surgeon and the experienced general surgeon as teacher/supervisor. In 477 (84 %) of all cases, preoperative localizations techniques were used—ultrasound in 457 patients (81 %), Sestamibi 399 (70 %), CT-scan 3 (0.5 %), PET 3 (0.5 %) and MRI 2 (0.4 %).
123
All surgery was done under general anesthesia and as inhospital procedures. Frozen-section histology was used in 505 (89 %) of cases. SPECT-CT and intraoperative parathyroid hormone measurements (IOPTH) were not used during this period. Follow-up was made with clinical examination including routine laryngoscopy, as well as measurement of calcium levels after 6 weeks and 6 months. At the 6 months follow-up, calcium status was recorded in 549 patients (96.8 %). Thyroid surgery (unilateral resection or lobectomy) was performed during parathyroid surgery in 63 patients (11 %), whereas thymus was partly or completely resected in 30 cases (5 %). Sternotomy was not necessary in any cases. Data were statistically analyzed in the medical database and analysis program ‘‘Medlog 2008-2’’ and in Microsoft Excel. v2-distribution test was used. Odds ratio (OR) was interpreted as the relative risk (RR).
Results BCE was performed in 287 cases (51 %), focused operation (visualization of only one gland) in 192 patients (34 %) and unilateral exploration (visualization of two glands unilaterally) in 88 (16 %) of cases. In 263 (92 %) of the BCEs, the procedure was planned preoperatively while 24 (8 %) were converted from focused or unilateral operations. In 498 patients (90.7 %), normocalcaemia was registered at follow-up blood testing, while 51 (9.3 %) had spontaneous hypercalcaemia interpreted as persistent disease. Complications Reoperation for postoperative hemorrhage was necessary in 6 patients (1.1 %) and postoperative wound infection was found in 6 patients (1.1 %). Preoperative laryngoscopy was performed in 563 patients (99.3 %). Combining follow-up data from the visits after 6 weeks and 6 months postoperative laryngoscopy was registered in 562 patients (99.1 %). In 6 patients (1.1 %), unilateral recurrent nerve paralysis caused by the operation was found. In 5 cases, no laryngoscopy was performed. No cases of bilateral paresis were found. As 287 procedures were bilateral, a total of 854 nerves were at risk—resulting in 0.7 % recurrent nerve paralysis of nerves at risk. Table 1 shows the complication rates for the overall series and for different types of surgery. While there is a tendency towards more frequent postoperative
Eur Arch Otorhinolaryngol Table 1 Complication rates for focused, unilateral and bilateral operations
Focused
Patients
Hemorrhage
Post infection
Recurrent nerve paresis 1 (0.5 %) NS
192
2 (1.0 %) NS
1 (0.5 %) NS
Unilateral
88
1 (1.1 %) NS
0 (0 %) NS
0 (0 %) NS
Bilateral
287
3 (1.1 %) NS
5 (1.7 %) NS
5 (1.7 %) (0.9 %NAR) NS
Overall
567
6 (1.1 %)
6 (1.1 %)
6 (1.1 %) (0.7 %NAR) NS
RNP recurrent nerve paresis, NAR nerves at risk, NS not significant
infections and recurrent nerve paresis in bilateral surgery, there is no significant difference. Histology In 515 patients (91 %), only one parathyroid gland was removed. Figure 1 shows the distribution of excised parathyroid glands. The median weight of parathyroid tissue removed was 0.43 g (range 0.05–7.69 g). Frozen sections were correct in 499 cases (99 %) compared to final histology. Parathyroid adenoma unfortunately lacks a precise histopathological definition [6, 7]. In this study, the diagnosis of adenoma was based on the presence of adjacent rim of normal parathyroid cells in the resected gland or a per-operative macroscopic or biopsy verified identification of normal or suppressed parathyroid glands. Final histology showed parathyroid adenoma in 435 cases (77 %), hyperplasia in 100 cases (18 %) and 2 lipo-adenomas (0.4 %). In 30 (6 %) of the explorations, no parathyroid tissue was found. These cases were all BCEs and were not excluded from analysis. Follow-up of these 30 patients interestingly showed persistent disease/failure in 52 % (18/34) of cases. No parathyroid cancers were diagnosed. Preoperative localization During the beginning of the study period, preoperative diagnostic localization procedures were not performed on a routine basis. From April 2006, preoperative 493
Paents
30
16 0 glands
1
2
7 3
18 Other
Fig. 1 Number of excised glands—493 (87 %) had one parathyroid gland removed. In 30 cases final histology showed no parathyroid tissue removed
ultrasonography as well as 99mTc Sestamibi scintigraphy (Sestamibi) was implemented as routine for all pHPT patients. To analyze the value of these measures we divided the material into two subgroups—before and after the 1st of April 2006. As demonstrated in Table 2, the implementation of routine preoperative localization procedures increased the rate of focused procedures by ten times and the rate of unilateral operations by 2.5 times. The cure-rate for the period 2004–2006 was 94.4 % and from 2006 to 2010 it was 88.7 %. The difference was not significant. Overall, during the 6-year period, the Sestamibi localized single gland disease correctly in 75 % of patients and ultrasound correctly in 61 % as described in Table 3. In 214 (37 %) cases, both ultrasonography and Sestamibi were correct and in this group, 97 % were cured by primary surgery. Risk factors for failure Table 4 shows univariate analysis of selected variables concerning risk of failure defined as hypercalcaemia 6 months postoperatively. Histology of hyperplasia was statistically significant with an odds ratio of 4.3 and p value \0.001. No other significant predictors of failure were found. A strong tendency towards failure was found in a small number of 29 patients with previous parathyroid surgery. In 6/29 (20.7 %) of cases, failure was found resulting in an OR of 2.5 (ns) (Table 4). In an attempt to preoperatively identify cases of hyperplasia—the patients with high risk of failure—further analysis was made. Table 5 shows analysis of factors related to hyperplasia. However, we did not find any significant preoperative factors to identify patients with hyperplasia. Learning setup The 50 patients who had surgery performed in a learning setup including experienced general surgeon as teacher and a younger surgeon as learner were identified retrospectively. None of these had recurrent nerve palsy or
123
Eur Arch Otorhinolaryngol Table 2 A remarkable increase of focused and unilateral operations followed the implementation of routine preoperative ultrasound and Sestamibi. Overall cure-rate was 90.7 %
Jan 2004–Mar 2006
Apr 2006–Dec 2009
Patients
197
370
Ultrasound
49 % (97)
99.5 % (368)
SM-scint.
19 % (38)
98 % (363)
US ? SM
8 % (15)
98 % (361)
Bilat expl
88 % (173)
31 % (114)
Unilat expl
8 % (15)
20 % (73)
Focused
5 % (9)
50 % (183)
Cure-rate (NS)
94.3 % (183/194) 3 unknown
88.7 % (314/354) 16 unknown
Reoperations
5 % (10)
5 % (19)
SM Sestamibi scintigraphy, US ultrasound, NS not significant
Table 3 Accuracy of preoperative ultrasound and Sestamibi scintigraphy localization diagnosis Localization
Sestamibi
Ultrasound
Correct Partially correct (multigland disease)
77 % (307) 0.5 % (2)
61 % (279) 1 % (5)
Inconclusive
17 % (68)
33 % (151)
6 % (22)
5 % (22)
False
In 214 (37 %) cases, both ultrasonography and Sestamibi were correct—in this group, 97 % were cured by primary surgery
postoperative infection. One patient (2 %) had a postoperative hemorrhage in need of acute reoperation. The lower complication rate in the learning setup patients was not significant. In three of the 50 ‘‘learningpatients’’, hypercalcaemia was found 6 months after surgery—resulting in a 94 % cure-rate.
Discussion Our introduction of Sestamibi and ultrasonography as standard preoperative localization examinations increased the use of unilateral explorations by 2.5 times and the use of focused surgery by 10 times. Though the duration of surgery was not measured in our study, the subsequent change towards unilateral and focused surgery has clearly decreased the duration of the operations and has allowed us to operate more patients per day. Cure rate and complications The 567 patients with primary hyperthyroidism were offered an effective and low risk treatment—overall, surgery cured 90.7 % of cases with low complication rates (Table 1). The change from routine BCE exploration to unilateral and focused surgery was, however, followed by
123
an unsatisfactory fall in cure-rate from 94.3 % before 2006 to 88.7 % from 2006. Though not significant, the fall in cure-rate after our introduction of focused surgery indicates that an increased focus on potential hyperplasia cases has to be introduced and that the threshold for conversion to BCE has to be low. A tendency towards lower complications rates was observed in unilateral and focused procedures but no significant difference was found, presumably due to a small number of patients (type II error). Focused and unilateral surgery is generally accepted as having lower morbidity and is widely considered to be the gold standard of parathyroid surgery today [4, 8]. Our results, however, show that hyperplasia (multigland disease) is a significant predictor of failure to cure the patients and that minimally invasive focused surgery may lead to increased risk of failure due to missed cases of hyperplasia. The high failure rate in patients where final histology showed no parathyroid tissue removed underlines the importance of correct preoperative localization and meticulous dissection—and emphasizes the potential of intraoperative PTH measurement to let the surgeon know when to stop the surgery. Literature In order to evaluate our results we selected comparable recent studies—an overview hereof is found in Table 6. Very high cure rates have been published from single institutions. Routine IOPTH seems to be a valuable tool to intraoperatively identify hyperplasia/multigland patients in need of BCE and thereby to increase cure rates. This has also been reported earlier by Gill et al. [9]. Bergenfelz et al. [5] have published a large multicenter Scandinavian study of 2708 patients to which we have also contributed. IOPTH was used in 29 % of cases and was found to significantly increase the chance of cure. In 2007, Allendorf et al. [10] published a single institution, one surgeon 17-year material of 1112 patients. All patients had BCE—in fact, 76 % of the procedures were performed in local anesthesia.
Eur Arch Otorhinolaryngol Table 4 Univariate analysis of selected variables concerning risk of failure Variable Age Gender Previous parathyroid surgery Type of exploration
Dichotomization
Observations
Failures
Percentage
p value
Odds ratio
NS
1.2
NS
1.7
NS
2.5
NS
1.10
NS
2.7
\0.001
4.3
NS
0.8
NS
1.4
NS
0.6
NS
1.0
NS
0.9
NS
1.1
Percentage
p value
Odds ratio
NS
0.8
NS
0.7
NS
1.0
NS
1.5
NS
1.0
NS
0.7
C60 years
282
28
9.9
\60 years
284
23
8.1
Male
121
16
13.2
Female
445
35
7.9
Yes
29
6
20.7
No
537
45
8.4
Unilateral
279
27
9.4
Bilateral
287
24
8.6
4
23.5
Hereditary endocrine disease
Yes
17
No
549
47
8.6
Histology
Hyperplasia Adenoma
99 434
16 16
16.2 3.7
Teaching Setup
Yes
52
4
7.7
No
514
47
9.1
Yes
476
45
9.5
No
90
6
6.7
Yes
329
23
7.0
No
237
28
11.8
Yes
274
25
9.1
No
292
26
8.9
Yes
138
11
8.0
No
428
40
9.4
Localization examinations preop. Symptomatic disease Ca?? level preoperatively above median (1.49) Ca?? level preoperatively [1.56 (upper quartile) Ca?? level preoperatively \1.44 (lower quartile)
Yes
157
15
9.6
No
409
36
8.8
Failure was defined as hypercalcaemia 6 months postoperatively
Table 5 Analysis of factors related to hyperplasia Variable Age Gender Symptomatic disease Preoperative calcium above median (1.49) Ca?? level preoperatively [1.56 (upper quartile) Ca?? level preoperatively \1.44 (lower quartile)
Dichotomization
Observations
Hyperplasia
C60 years
282
44
15.6
\60 years
285
56
19.7
Male
445
71
16.0
Female
122
29
23.8
Yes
330
58
17.6
No
237
42
17.7
Yes
274
59
21.5
No
293
41
14.0
Yes
138
25
18.1
No
429
75
17.5
Yes No
157 410
20 80
12.7 19.5
No significant factors were found
Siperstein et al. [11] found that 16 % of cases of multigland disease were overseen due to limitations of localization techniques. Lew et al. [12] analyzed this further and
concluded that inability of the surgeon rather than multigland disease was the reason for failure—at least in the presence of IOPTH. In a large study of 15,000 parathyroid
123
Eur Arch Otorhinolaryngol Table 6 Overview of comparable current studies of surgery for pHPT Author
Patients/period
Cure-rate (%)
Complication rates
Remarks
Bergenfelz et al. [5]
2708 (2004–2008)
93
0.8 % postop. bleeding, 1.2 % infection and 1.4 % unilateral recurrent nerve palsy
Multicenter, Scandinavia. IoPTH in 29 % of cases
Allendorf et al. [10]
1112 (1987–2003)
97.4
0.2 % recurrent nerve injury and postop. bleeding of 0.8 %
Single institution, New York. Single surgeon. 76 % in local anesthesia
Siperstein et al. [11]
1158 (1999–2007)
98
Not described
Single institution, Ohio. All bilateral explorations, IOPTH in use
Lew et al. [12]
845 (1993–2009)
97.1
Not described
Single institution, Miami. All IOPTH
Lee et al. [14]
538 (1998–2010)
96.6
Not described
Single institution, Texas, All IOPTH
Present study
567 (2004–2009)
90.7
1.1 % for bleeding, 1.1 % for wound infection and 0.9 % for recurrent nerve paralysis
Single institution, Denmark. IOPTH not in use. 99 % laryngoscopy pre and postop.
operations, Norman et al. [13] had similar findings and argued for a return to routine BCE to reduce failures. In a recent review of 538 patients, Lee et al. [14] found a cure-rate of 96.6 % with focused surgery. In a systematic analysis of their failures, they found multigland disease, falsely positive IOPTH levels, and surgeon’s inability to identify the abnormal parathyroid glands to be the causes of their failures. Their advice was to consider conversion to BCE in selected cases with low IOPTH level drops.
Conclusions No increase in complication rates was detected during the time period in which ENT head and neck surgeons were trained in parathyroid surgery, indicating that a shift between specialties is possible without any negative impact on surgical quality. Routine presurgical diagnostics makes focused surgery feasible but the price might be an increased failure rate—primarily because of not diagnosed parathyroid hyperplasia. Conflict of interest of interest.
The authors declare that they have no conflict
Ethical standard All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent Informed consent was obtained from all individual participants included in the study.
References 1. Fraser WD (2009) Hyperparathyroidism. Lancet 374(9684): 145–158. doi:10.1016/S0140-6736(09)60507-9
123
2. Babinska D, Barczynski M, Stefaniak T, Oseka T, Babinska A, Babinski D, Sworczak K, Lachinski AJ, Nowak W, Sledzinski Z (2012) Evaluation of selected cognitive functions before and after surgery for primary hyperparathyroidism. Langenbecks Arch Surg 397(5):825–831. doi:10.1007/s00423-011-0885-5 3. Edwards ME, Rotramel A, Beyer T, Gaffud MJ, Djuricin G, Loviscek K, Solorzano CC, Prinz RA (2006) Improvement in the health-related quality-of-life symptoms of hyperparathyroidism is durable on long-term follow-up. Surgery 140(4):655–663. doi:10. 1016/j.surg.2006.06.016 (discussion 653–654) 4. Udelsman R, Lin Z, Donovan P (2011) The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism. Ann Surg 253(3):585–591. doi:10.1097/SLA.0b013e318208fed9 5. Bergenfelz AO, Jansson SK, Wallin GK, Martensson HG, Rasmussen L, Eriksson HL, Reihner EI (2009) Impact of modern techniques on short-term outcome after surgery for primary hyperparathyroidism: a multicenter study comprising 2,708 patients. Langenbecks Arch Surg 394(5):851–860. doi:10.1007/ s00423-009-0540-6 6. Ghandur-Mnaymneh L, Kimura N (1984) The parathyroid adenoma. A histopathologic definition with a study of 172 cases of primary hyperparathyroidism. Am J Pathol 115(1):70–83 7. Erickson LA, Jin L, Wollan P, Thompson GB, van Heerden JA, Lloyd RV (1999) Parathyroid hyperplasia, adenomas, and carcinomas: differential expression of p27Kip1 protein. Am J Surg Pathol 23(3):288–295 8. Kunstman JW, Udelsman R (2012) Superiority of minimally invasive parathyroidectomy. Adv Surg 46:171–189 9. Gill MT, Dean M, Karr J, Aultman DF, Nathan CO (2011) Intraoperative parathyroid hormone assay: a necessary tool for multiglandular disease. Otolaryngol Head Neck Surg 144(5):691–697. doi:10.1177/0194599811398597 10. Allendorf J, DiGorgi M, Spanknebel K, Inabnet W, Chabot J, Logerfo P (2007) 1112 consecutive bilateral neck explorations for primary hyperparathyroidism. World J Surg 31(11):2075–2080. doi:10.1007/s00268-007-9068-5 11. Siperstein A, Berber E, Barbosa GF, Tsinberg M, Greene AB, Mitchell J, Milas M (2008) Predicting the success of limited exploration for primary hyperparathyroidism using ultrasound, sestamibi, and intraoperative parathyroid hormone: analysis of 1158 cases. Ann Surg 248(3):420–428. doi:10.1097/SLA. 0b013e3181859f71 12. Lew JI, Rivera M, Irvin GL 3rd, Solorzano CC (2010) Operative failure in the era of focused parathyroidectomy: a contemporary
Eur Arch Otorhinolaryngol series of 845 patients. Arch Surg 145(7):628–633. doi:10.1001/ archsurg.2010.104 13. Norman J, Lopez J, Politz D (2012) Abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid operations. J Am Coll Surg 214(3):260–269. doi:10. 1016/j.jamcollsurg.2011.12.007
14. Lee S, Ryu H, Morris LF, Grubbs EG, Lee JE, Harun N, Feng L, Perrier ND (2014) Operative failure in minimally invasive parathyroidectomy utilizing an intraoperative parathyroid hormone assay. Ann Surg Oncol. doi:10.1245/s10434-013-3479-3
123
