ORIGINAL COMMUNICATION
Experience in Identifying the Venous Drainage of the Adrenal Gland During Laparoscopic Adrenalectomy
CRAIG N. PARNABY,1* NORMAN GALBRAITH,2 AND PATRICK J. O’DWYER1
1Department of Surgery, Western Infirmary, Glasgow, Scotland, United Kingdom
2Department of Anatomy, University of Glasgow, Glasgow, Scotland, United Kingdom
Laparoscopic adrenalectomy (LA) is the procedure of choice for most adrenal tumors. An important part of LA is the early identification and ligation of the adrenal veins. The venous drainage from each adrenal gland is usually via a single vein: the right vein draining into the inferior vena cava (IVC) and the left vein into the left renal vein. Although infrequent, variable venous drainage has been documented. The aim of the study was to clarify if LA identified ve- nous drainage and its variants. Between January 1999 and January 2008, 142 consecutive patients underwent LA. Adrenal vein anatomy was documented on a prospective database. In total, 142 patients underwent 162 LA (right = 62, left = 66, bilateral = 17). All adrenal veins were identified at the time of lapa- roscopy. For 157 LA, the adrenal venous drainage was constant: right vein drained into the IVC and left vein drained into left renal vein. Five patients had adrenal vein variants: two right veins draining separately into IVC (n = 1), two right veins draining into the IVC and right renal vein (n = 1), and two left veins draining separately into the left renal vein (n = 3). Adrenal vein variants were present in patients with phaeochromocytomas (n = 4) or adrenocortical carci- noma (n = 1). The laparoscopic approach allowed an excellent view of the main adrenal venous anatomy. This has helped confirm the constant nature of the venous drainage and successfully identify variant adrenal veins. Clin. Anat. 21:660-665, 2008. @2008 Wiley-Liss, Inc.
Key words: laparoscopic adrenalectomy; adrenal vein; adrenal artery; anatom- ical variations; pathology
INTRODUCTION
This article does not give a comprehensive account of the adrenal veins but our surgical experi- ence of them in a large laparoscopic adrenalectomy (LA) series. The venous drainage from each adrenal gland, described in standard anatomical textbooks, is usually via a single vein: the right joins the inferior vena cava (IVC) and the left one joins the left renal vein (Standring, 2005). Variations to this pattern have been documented in a number of cadaver stud- ies (Table 1). Anson et al. (1948), in a study of 450 cadavers, confirmed the constant venous drainage. It was not clear from this study of kidney and adre- nal gland blood supply, if the main adrenal vein was
dissected in all cadavers or if a more selective policy was used. Results of subsequent cadaver studies which focused on the venous drainage in normal ad- renal glands, however, described increased variabili- ty of the main venous drainage (particularly on the right side) and a network of veins surrounding the
*Correspondence to: Craig N. Parnaby, Department of Surgery, Western Infirmary, Glasgow, G11 6NT, Scotland, United Kingdom. E-mail: craigparnaby@aol.com
Received 28 March 2008; Revised 9 July 2008; Accepted 30 July 2008
Published online 23 September 2008 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ca.20706
| Author (year) | Number of cadavers | Standard venous anatomy right side (%) | Variable venous anatomy right side (%) | Standard venous anatomy left side (%) | Variable venous anatomy left side (%) |
|---|---|---|---|---|---|
| Anson et al. (1948) | 450 | 450 (100) | 0 | 449 (99.8) | 1 (0.2) |
| Johnstone (1957) | 10 | 5 (50) | 5 (50) | 9 (90) | 1 (10) |
| Clark (1959) | 16 | 6 (38) | 10 (62) | 11 (69) | 5 (31) |
| El-Sherief (1982) | 20 | 9 (45) | 11 (55) | 20 (100) | 0 |
| Monkhouse and Khalique (1986)ª | 57 | 27 (60) | 18 (40) | 57 (100) | 0 |
ªTwelve right adrenal glands not suitable for dissection.
adrenal gland (Johnstone, 1957; Clark, 1959; El- Sherief, 1982; Monkhouse and Khalique, 1986). The variability in terms of duplication, triplication, and the termination of the main adrenal veins were described in these studies. The arterial supply of the adrenal glands has been well documented. Anson et al. (1948) described the arterial supply of the adre- nal glands, in that they can receive up to 60 rami (small arteries) derived from aorta, inferior phrenic, and renal arteries.
Since it was first introduced in 1992, LA (Gagner et al., 1992) has shown benefits in terms of safety, recovery, analgesic requirements, and hospital stay when compared with the open procedure (Brunt et al., 1996; Dudley and Harrison, 1999). LA has become the procedure of choice for most adrenal pathologies. During LA, early identification and liga- tion of the main adrenal vein is an important step (Gagner et al., 1997). This is mainly to help orienta- tion and prevent excessive hemorrhage.
The cadaver studies aforementioned were carried out on nondiseased adrenal glands. It is not known if adrenal tumors, because of angiogenesis or vasodila- tation of preexisting vessels, may increase both the variability of venous drainage and the number of periadrenal vessels. The purpose of this study was to document the surgical importance of variable adrenal vein anatomy and the multiple periadrenal vessels during LA.
MATERIALS AND METHODS
Details of all patients referred to the University Department of Surgery with adrenal tumors between January 1999 and January 2008 had been recorded prospectively on a database. Details of the main ad- renal veins and any periadrenal vessel variants were recorded in each operation note.
All patients were assessed preoperatively by the endocrine unit. All adrenal tumors were screened for hormone excess. A phaeochromocytoma screen included urinary catecholamines (epinephrine, nor- epinephrine, or dopamine) and metabolites (methoxy-derivatives or vanillyl mandelic acid). A Conn’s syndrome screen included urea and electro- lytes, plasma renin, and aldosterone levels. Adrenal vein sampling was used selectively in these patients
when the initial assessment did not give a clear diagnosis. Cushing’s syndrome/disease assessment included 24-hr urinary cortisol, serum cortisol, and plasma ACTH, low dose dexamethasone suppression test, and plasma androgen measurements.
All adrenalectomies were carried out by the same surgeon. The only contraindications to LA were locally invasive adrenal carcinoma or the require- ment of an additional open surgical procedure. LA was performed using the lateral transabdominal approach as described by Gagner et al. (1992). Dur- ing laparoscopy, the main vein draining the tumor was identified and ligated using liga-clips. Other smaller periadrenal vessels were divided using elec- trocautery or ultrasonic dissection (use of ultrasound to cut or coagulate vessels).
Preoperative information included patient demo- graphics and diagnosis. Intraoperative information included details on the main venous drainage, any variant arterial or venous anatomy, and any bleeding complications. Postoperative data included any fur- ther bleeding complications and the requirement for intervention.
| All patients | ||
|---|---|---|
| Number of patients | 142 | |
| Number of laparoscopic adrenalectomies | 162 | |
| Sex ratio (M:F) | 55:87 | |
| Age (years)ª | 56 (25-83) | |
| Tumor siteb | ||
| Right | 62 | |
| Left | 66 | |
| Bilateral | 17 | |
| Indication for referralb | ||
| Conn's adenoma | 36 | |
| Phaeochromocytoma | 40 | |
| Cushing's syndrome | 25 | |
| Cushing's disease | 9 | |
| Nonfunctioning adrenal tumor | 7 | |
| Possible adrenal carcinoma | 18 | |
| Possible adrenal metastasis | 9 | |
| Mixed secreting adenoma | 1 | |
ªValues are mean (range).
bThree patients had asynchronous adrenalectomy because of further adrenal gland pathology [Cushing’s syndrome (n = 2), phaeochromocytoma (n = 1)].
| Patients | Side | Pathologyª | Size (mm) | Type of variability | Blood loss (ml) | Type of bleeding problems |
|---|---|---|---|---|---|---|
| 1 | Right | Phaeo | 80 | 2 adrenal veins into IVC | 200 | Large tumor (8 cm) ooze from periadrenal vessels |
| 2 | Left | Phaeo | 30 | 1 adrenal vein into left renal vein 1 adrenal vein unclear site of drainage | Minimal | None |
| 3 | Left | Phaeo | 80 | 2 adrenal veins into left renal vein | Minimal | None |
| 4 | Left | Phaeo | 55 | 2 adrenal veins into left renal vein | Minimal | None |
| 5 | Right | ACC | 150 | 1 adrenal vein into IVC 1 adrenal vein into right renal vein | Minimal | None |
ªACC, adrenocortical carcinoma; phaeo, phaeochromocytoma.
RESULTS
Over the 9-year period, 162 LA were carried out on 142 consecutive patients. Characteristics of the patients are shown in Table 2. Of the 162 LAs, the main adrenal vein was identified in all cases at time of laparoscopy. For 157 LAs, the main venous drainage was constant via a single vein (the right adrenal vein into the IVC and the left adrenal vein into the left renal vein). For the other five patients, the main venous drainage was variable. Character- istics of the patients with a variable venous pattern are shown in Table 3. There were no bleeding com- plications due to failure to identify or ligate the ad- renal vein variants. Of the five patients, only one had measurable blood loss (200 ml). This patient had a large (8 cm) phaeochromocytoma with per- sistent ooze from the multiple periadrenal vessels. Conversion to an open procedure was carried out because of concern over adjacent structure inva- sion. There were no other converted procedures for patients with variable adrenal vein drainage. The patient with a right adrenocortical carcinoma had a prominent, dilated second adrenal vein draining into the right renal vein. This has been illustrated by a diagram (Fig. 1) and magnetic resonance image (Fig. 2).
The characteristics of phaeochromocytomas with variable venous drainage compared with those with standard venous drainage are shown in Table 4. Both groups secreted a range of catecholamines. The median size of the phaeochromocytomas with adre- nal vein variants was 67 mm (IQR 36-80) compared with a median of 55 mm (IQR 35-70) for those with standard venous drainage.
During laparoscopy, apart from the main adrenal vein, a high number of other small vessels were always present around the adrenal gland. It was not possible to differentiate between veins and arteries. These were all dealt with in the same way, by dia- thermy or ultrasonic dissection. Because of the small size of these vessels there was no requirement for other haemostatic techniques. No laparoscopic pro- cedure required conversion because of bleeding from these smaller vessels. For four patients with large (>6 cm) phaeochromocytomas (n = 3) and a large (>6 cm) adrenal metastatic deposit secondary to re-
nal cell carcinoma (n = 1) there were increased num- bers of periadrenal vessels. These were all divided without incident by careful ultrasonic dissection. The main venous drainage was by a single vein for these patients. In one patient with a phaeochromocytoma, the patient appeared initially to have a duplicated left adrenal vein. As dissection progressed, however, the second vein was found to travel into perinephric fat without obvious communication with the phaeo- chromocytoma.
Liver
Inferior vena cava
large adrenal tumour
adrenal vein
2nd adrenal vein
kidney
renal vein
MRI adrenal to illustrate duplicate adrenal vein for a patient with a large adrenal tumour
Adrenal vein
Large adrenal tumour
Inferior vena cava
Kidney Renal Vein 2nd adrenal vein
DISCUSSION
The majority of patients (97%) undergoing LA had constant venous drainage. This was via a single vein draining into the IVC on the right and draining into left renal vein on the left. When variation did occur, the main vein was duplicated. Other multiple, small periadrenal vessels were noted during laparoscopy for all patients. It was not possible to distinguish between small veins and arteries. Small periadrenal veins may provide alternative venous drainage path- ways. In this study, however, they were of no surgi- cal significance and were all divided easily using electrocautery or ultrasonic dissection.
Laparoscopy allows an excellent, magnified view (2.5-3 times normal) of the adrenal gland and sur- rounding structures compared with the traditional view seen at open surgery. Despite these advan- tages, it is not known if laparoscopy allows an accu- rate view of the venous anatomy. Identification of the main adrenal vein is an essential part of LA (Gagner et al., 1997). This study has demonstrated the main venous drainage (including variants) can be clearly defined at time of LA for all patients. No
bleeding complications or the requirement to convert to an open procedure resulted from failure to recog- nize the main adrenal vein and its variants.
A number of series have dissected adrenal veins in cadavers (Anson et al., 1948; Johnstone, 1957; Clark, 1959; El-Sherief, 1982; Monkhouse and Khali- que, 1986). In all these series there was no adrenal gland pathology. In a large series of 450 cadavers, Anson et al. (1948) looked at the pararenal system of veins and arteries and commented on the con- stant nature of the venous drainage of the adrenal gland via a single vein. Only one variation occurred, where the left adrenal vein joined the right renal vein. There was no mention of duplication or triplica- tion of the adrenal vein. It is not clear, however, if the authors carried out selected adrenal vein dissec- tion as the study did not specifically examine the ve- nous drainage of the adrenal gland. There was also no mention of the network of smaller veins sur- rounding the adrenal gland. Thus, this description of adrenal gland venous drainage should be interpreted with caution.
Subsequent series focused on venous drainage and highlighted its variable nature, particularly on
| Characteristic | Variable venous drainage | Standard venous drainage |
|---|---|---|
| Type of secretion | ||
| Epinephrine | 1 | 7 |
| Norepinephrine | 1 | 9 |
| Dopamine | 0 | 1 |
| Epinephrine/norepinephrine | 0 | 4 |
| Epinephrine/norepinephrine/ dopamine | 0 | 5 |
| No typical excretionª | 2 | 7 |
| No secretion | 0 | 3 |
| Size (maximum | 67 | 55 |
| diameter, mm)b | (36-80) | (35-70) |
aDefined as elevation of methoxy-derivatives or vanillyl mandelic acid.
bMedian (interquartile range).
the right side (Table 1). Compared with Anson et al. (1948), however, the numbers in these series were much smaller (Johnstone, 1957; Clark, 1959; El- Sherief, 1982; Monkhouse and Khalique, 1986). In these series, using fresh or embalmed cadaver ma- terial, variability on the right side was fairly con- stant, occurring in 40-60% of cases. The variability involved duplication or triplication of the main vein. In addition, alternative venous drainage occurred into the hepatic or renal veins rather than the IVC. Johnstone (1957), in 3 of 10 cadaver dissections, described the right adrenal vein draining directly into a right hepatic vein close to its junction with the IVC. Variability on the left side appeared to be infrequent in the majority of series. Johnstone (1957) and Clark (1959) commented on a duplicated left adrenal vein. Clark (1959) described 5 of 16 cadavers having duplication of the main vein with both draining into the left renal vein and Johnstone (1957) reported that 1 of 10 dissections had a second main vein draining into a left lumbar vein.
In the current series a variable main adrenal vein was infrequent (5 of 162 adenalectomies). The single main left adrenal vein is in agreement with the majority of other studies. The variations we saw were all duplications of the adrenal vein. Duplicated veins have been described previously by Johnstone (1957), Clark (1959), El-Sherief (1982), and Monk- house and Khalique (1986).
The presence of multiple periadrenal vessels has been well described in a number of anatomical stud- ies. Johnstone (1957), in a series of 10 cadaver dis- sections, described the drainage pattern of multiple venules surrounding the adrenal gland. The venules could drain into the main adrenal vein, the veins of the perirenal fat, renal vein, inferior phrenic vein, or lumbar veins. Anson et al. (1948) described the abundant arterial supply of the adrenal glands. They demonstrated that up to 60 branches derived from the inferior phrenic, aorta, or renal arteries can sup- ply the adrenal gland. Folkman (1995) described the
necessity of angiogenesis to allow tumor growth and subsequent metastasis. In the presence of an adre- nal tumor, new periadrenal vessels may sprout from preexisting vessels and lead to troublesome bleeding at LA. These changes would be expected to be more apparent as the tumor increases in size. Our study, demonstrated the abundance of vessels around the adrenal gland. The periadrenal vessels did not pres- ent a problem in the majority of cases. The vessels were controlled with electrocautery or ultrasonic dis- section. For three patients with large phaeochromo- cytomas (≥6 cm) and one patient with a large adre- nal metastasis, however, there were an increased number of periadrenal vessels. Another patient with a large adrenocortical carcinoma (15 cm) had multiple prominent vessels, particularly a tortuous, dilated second adrenal vein draining into the right renal vein. A possible explanation could be compres- sion of the main adrenal vein (because of the mass effect of the tumor) which caused preferential ve- nous return down the second adrenal vein.
In this study, the adrenal vein variants all occurred in patients with phaeochromocytomas or a large (15 cm) adrenocortical carcinoma. Of 42 LA for phaeochromocytomas (two bilateral), only four showed variant venous anatomy. The range of cate- cholamine secretion was comparable to phaeochro- mocytomas with standard venous anatomy. This is the first study to describe main adrenal vein variants during LA for a series of patients with phaeochromo- cytomas. Other series have described only standard venous anatomy during LA for phaeochromocytoma (Cheah et al., 2002; Gotoh et al., 2002).
Our study has demonstrated that LA can accu- rately visualize the main venous drainage (and any variants) from the adrenal gland with minimal bleed- ing complications. This has helped confirm the con- stant nature of the main adrenal veins in the major- ity of patients undergoing LA. However, adrenal vein variants and an increased number of periadrenal vessels appeared to occur in patients with phaeo- chromocytomas or large adrenocortical carcinomas. Increased attention to the adrenal gland vasculature may be required during LA in these patients.
REFERENCES
Anson BJ, Cauldwell EW, Pick JW, Beaton LE. 1948. The anatomy of the pararenal system of veins, with comments on the renal arteries. J Urol 60:714-737.
Brunt LM, Doherty GM, Norton JA, Soper NJ, Quasebarth MA, Moley JF. 1996. Laparoscopic adrenalectomy compared to open adre- nalectomy for benign adrenal neoplasms. J Am Coll Surg 183:1- 10.
Cheah WK, Clark OH, Horn JK, Siperstein AE, Duh QY. 2002. Lapa- roscopic adrenalectomy for pheochromocytoma. World J Surg 26:1048-1051.
Clark K. 1959. The blood vessels of the adrenal gland. J R Coll Surg Edinb 4:257-262.
Dudley NE, Harrison BJ. 1999. Comparison of open posterior versus transperitoneal laparoscopic adrenalectomy. Br J Surg 86:656- 660.
El-Sherief. 1982. Adrenal vein catheterization. Anatomic considera- tions. Acta Radiol Diagn (Stockh) 23:345-360.
Folkman J. 1995. Seminars in medicine of the Beth Israel Hospital, Boston. Clinical applications of research on angiogenesis. N Engl J Med 333:1757-1763.
Gagner M, Lacroix A, Bolte E. 1992. Laparoscopic adrenalectomy in Cushing’s syndrome and pheochromocytoma. N Eng J Med 327:1033.
Gagner M, Pomp A, Heniford BT, Pharand D, Lacroix A. 1997. Lapa- roscopic adrenalectomy: Lessons learned from 100 consecutive procedures. Ann Surg 226:238-247.
Gotoh M, Ono Y, Hattori R, Kinukawa T, Ohshima S. 2002. Laparo- scopic adrenalectomy for pheochromocytoma: Morbidity com- pared with adrenalectomy for tumors of other pathology. J Endourol 16:245-249.
Johnstone FR. 1957. The suprarenal veins. Am J Surg 94:615-620. Monkhouse WS, Khalique A. 1986. The adrenal and renal veins of man and their connections with azygos and lumbar veins. J Anat 146:105-115.
Standring S. 2005. Suprarenal (adrenal) gland. In: Standring S. (ed.) Gray’s Anatomy. 39th Ed. London: Elsevier. p 1245-1249.