Long-term safety of left renal vein division and ligation to expedite complex abdominal aortic surgery
Article Outline
Background
Left renal vein division and ligation (LRVDAL) is performed to facilitate complex abdominal aortic surgery. Surgeons restore continuity of the vein due to concern that ligation could cause renal compromise or hematuria. However, we report the short and long-term safety of left renal vein division and ligation.
Method
Between 1992 and 2007, we divided the left renal vein in 56 patients (40 males, 16 females) ages 57 to 84 (average 74-years-old) who were treated for aortic occlusive disease (9) or abdominal aortic aneurysm (47). Patients requiring concomitant renal artery reconstruction were excluded from this review. Suprarenal cross-clamp was used in 51 patients with temporary vessel-loop control of the renal arteries. Creatinine (Cr) and glomerular filtration rates (eGFR) were measured pre-, post-, and long-term after surgery. Outpatient records of all patients that had survived more than 12 months were also reviewed in order to evaluate the late effects on renal function or symptoms possibly related to LRVDAL.
Results
Median procedure duration was 157 (61-375) minutes. Median cross-clamp time was 16 (10-45) minutes. Median intensive care unit (ICU) and hospital length of stays were 2 (1-11) days and 7 (4-58) days, respectively. There were no deaths. There were no complications directly related to renal vein ligation. Hematuria, seen in 2 patients, was a result of traumatic insertion of a Foley catheter. Median pre-op and discharge Cr levels were 1.1 mg/dL (0.7-2.4 mg/dL) and 1.1 mg/dL (0.6-2.1 mg/dL), respectively (P < .5). Median change in Cr was 0.0 mg/dL and only increased in 14 patients (maximum increase 0.9 mg/dL). Median pre-op and discharge eGFR was 61 mL/minute (28-137 mL/minute/1.73 m2) and 67 mL/minute (32-138 mL/minute/1.73 m2), respectively (P < .5). Cr and eGFR in the 2 patients with a Cr of >2.0 mg/dL remained unchanged post-op. Only 2 patients with a Cr of <2.0 mg/dL had a post-op Cr >2.0 mg/dL and both returned to normal by day 3 post-op. Thirty-six patients have been followed for more than a year (median 34.5 months, maximum 144 months) and Cr has remained stable in all but 2 patients. These 2 patients, both with a pre-op Cr of 1.5 mg/dL, subsequently developed Cr levels of 2.1 mg/dL and 2.4 mg/dL but maintained baseline Cr levels for 25 and 34 months, respectively, before demonstrating these elevated levels which have proven to be unrelated to renal vein ligation. Hematuria and flank pain have never been recorded after discharge.
Conclusion
Restoration of left renal vein continuity after LRVDAL may be unnecessary since renal compromise and hematuria was not encountered in this long-term analysis.
Endovascular techniques have allowed the vascular surgeon to treat most patients whose aortic occlusive or aneurysmal disease is associated with favorable anatomy. During this paradigm shift, we, as well as others, have seen an increase in the complexity of open aortic reconstruction1 due to heavily calcified aortas or those with wide, angled, short, or absent aortic necks. Accordingly, safe surgery will frequently now demand excellent exposure of the juxta-renal aorta. In order to better visualize this anatomic region, we have increasingly adopted the technique of left renal vein (LRV) ligation and division (left renal vein division and ligation [LRVDAL]) which was first described by Neal and Shearburn in 1967.2 Left renal vein division and subsequent re-anastomosis (LRVDAR) in order to facilitate such exposure was originally described by Szilagyi et al3 and this technique has been adopted by many surgeons. However, since then there have been reports attesting to the safety of LRVDAL although there have also been manuscripts suggesting that failure to reconstruct the left renal vein could lead to deterioration in renal function as well as other untoward complications.
In an attempt to determine whether LRVDAL has any potential adverse sequelae, we have performed a retrospective review of our experience. Our manuscript is unique in that it is the first to evaluate not only in-hospital creatinine (Cr) and estimated glomerular filtration rates (eGFR) levels but also the long term consequences of LRVDAL.
Materials and methods
From August 1992 through December 2007, 1012 patients underwent open aortic surgery. In order to prevent compounding variables, patients who underwent concomitant renal artery reconstruction were excluded from this review, except for 1 patient who had an aberrant secondary right renal artery re-anastomosed to the aortic graft at the time of the procedure. Fifty-six of these patients required concomitant LRVDAL in order to enable improved exposure of the juxtarenal aorta during open surgery for aortic occlusive disease (9) or abdominal aortic aneurysm (47). LRVDAR was never performed. All procedures were performed by one of three board certified vascular surgeons as the primary surgeon. Typically, two of the three surgeons worked together using a standard transabdominal/transperitoneal approach. The renal arteries were controlled with vessel loops to prevent atheroembolization whenever suprarenal cross-clamping was required (51 of 56 patients). The left renal vein was ligated and then suture ligated with 2.0 silk before being divided close to the inferior vena cava in an attempt to preserve collateral venous drainage to the left kidney. The major tributaries, ie, the gonadal, adrenal, and lumbar veins were preserved in all patients. Also, the right kidney and its renal vein were anatomically normal in all patients. All patients were heparinized prior to aortic cross-clamping, and reversed with protamine sulphate at the end of the procedure. Mannitol (12.5 grams) was given in all cases prior to aortic cross-clamping. Standard cell saver collection and reinfusion techniques were used in most cases. Aortic reconstruction was performed in all cases with woven polyester Dacron grafts.
A retrospective review of hospital records was performed with attention to operative reports, pre- and postoperative labs, and perioperative complications. Operative times and blood loss were recorded for all operations. Demographic data was also collected including: age, gender, tobacco use, coronary artery disease, hypercholesterolemia, diabetes, chronic obstructive pulmonary disease, peripheral arterial disease, hypertension, and renal insufficiency (Cr ≥1.5 gm/dL). Preoperative, in-hospital, and discharge Cr and eGFR were recorded. The eGFR was calculated using the Cockcroft-Gault equation4, 5 (140-age) × weight/72 × serum Cr (age is in years, body weight is in kilograms, and serum Cr is in mg/dL; in women, the equation is multiplied by 0.85). The eGFR values are expressed as mL/minute/1.73 m2. Outpatient records of all patients that had survived more than 12 months were also reviewed in order to evaluate late effects on renal function or symptoms possibly related to LRVDAL.
All in-hospital complications were recorded using the following criteria: blood transfusions that were required in the postoperative course, stroke, pulmonary embolism, or deep venous thrombosis as documented by computed tomographic angiography (CTA) scan and/or venous duplex ultrasound scan or d-dimer elevation when the diagnosis required laboratory confirmation, myocardial infarction as documented by electrocardiographic changes and enzymatic elevation, electrocardiogram (EKG) evidence of arrhythmias, congestive heart failure identified by discharge diagnosis and documented by cardiology consultation, hematuria confirmed by urine analysis, renal failure (Cr ≥2.0 mg/dL), ileus defined as inability to advance diet by postoperative day (POD) #3, respiratory complications if the patient could not be extubated within 24 hours, required reintubation, developed pneumonia, or required a tracheostomy. Hospital and intensive care unit (ICU) length of stay were also recorded.
Results
There were 56 patients (males 40 [71%], females 16 [29%]) reviewed for this study. Mean age was 74 years (range, 57-84 years). Comorbidities are listed in Table I. Significantly, coronary artery disease (CAD) was present in 36 (64%) patients and hypertension was prevalent (68%). Baseline renal insufficiency (Cr ≥1.5 mg/dL) was noted in 8 (14%) of the patients.
Table I. Comorbidities encountered in the 56 patients undergoing left renal vein division and ligation
| Comorbidities | Patients, n (%) |
|---|---|
| CAD | 36 (64%) |
| Hypertension | 38(68%) |
| Tobacco | 28(56%) |
| Hypercholesterolemia | 31(55%) |
| Peripheral arterial disease | 35(63%) |
| Diabetes | 10(18%) |
| Pulmonary disease | 22(39%) |
| Renal insufficiency (Cr ≥1.5 mg/dL) | 8(14%) |
Because of short aortic necks or heavy infrarenal aortic calcification, the majority of the patients (51 [91%]) underwent suprarenal aortic cross-clamping. Median suprarenal cross-clamp time was 16 minutes (range, 10-45). Median operative time was 157 minutes (range, 61-375). Median blood loss was 1125 mL (range, 250-3000 mL). Median ICU stay was 2 days (range, 1-11 days) with median overall length of stay 7 days (range, 4-58).
In-hospital morbidity and mortality
There were no deaths in our patient cohort (Table II). There were no strokes, pulmonary embolisms, or limbs lost. Seven (13%) total patients experienced delayed return of bowel function lasting >3 days postoperatively (range, 4-7 days). Cardiac complications occurred in 8 (14%) patients, arrhythmias 5 (9%), myocardial infarction 1 (2%), and congestive heart failure 2 (4%). Five patients (9%) experienced respiratory complications, but only 1 required a tracheostomy. Hematuria was noted in only 2 (4%) patients. Both of these patients had traumatic Foley catheter insertions and cleared their urine within 72 hours.
Table II. Complications following aortic surgery in the 56 patients with concomitant left renal vein division and ligation
| Complication | Patients, n (%) |
|---|---|
| Deaths | 0(0%) |
| Strokes | 0(0%) |
| Pulmonary embolism | 0(0%) |
| Limb loss | 0(0%) |
| Renal failure(Cr ≥2.0 mg/dL) | 1(2%) |
| Hematuria | 2(4%) |
| Respiratory | 5(9%) |
| Ileus | 7(13%) |
| Total cardiac | 8(14%) |
| Arrhythmias | 5(9%) |
| MI | 1(2%) |
| CHF | 2(4%) |
Cr and eGFR
Average and median preoperative Cr was 1.13 mg/dL and 1.1 mg/dL, respectively, (range, 0.7-2.4 mg/dL). Calculated preoperative average eGFR was 66.3 mL/minute/1.73 m2 with a median of 60.7 mL/minute/1.73 m2 (range, 28.0-137.0 mL/minute/1.73 m2). Median change in Cr was 0.0 mg/dL and only increased in 14 patients (maximum increase 0.9 mg/dL). Median Cr at discharge was similar to pre-op (1.1 mg/dL [0.6-2.1 mg/dL]). Median eGFR increased from 60.7 (range, 28.1-137.1) to 66.8 (32.1-138.0) mL/minute/1.73 m2 but this was not significant (P < .5) (see Table III and the Fig). Eight patients (14%) were considered to have renal insufficiency preoperatively defined as a Cr >1.5 mg/dL (median Cr 1.6 mg/dL; range, 1.5 to 2.4 mg/dL), yet 5 patients demonstrated a decrease in Cr and a concomitant increase in eGFR to normal levels postoperatively. Further, only 1 of these 8 patients had an increase in Cr (from 1.6 mg/dL to 2.3 mg/dL on POD #3). He was discharged home at POD #6 with a Cr of 1.8 mg/dL. The other 2 patients also had discharge Cr lower than preoperative values (1.6 mg/dL to 1.5 mg/dL and 2.4 mg/dL to 1.5 mg/dL). The Cr values for these 8 patient pre- and postoperatively were 1.6 to 1.5, 1.8 to 1.2, 1.5 to 1.0, 1.5 to 1.2, 1.6 to 1.1, 1.5 to 1.2, 2.4 to 1.5, and 1.6 to 1.8. The 3 patients with an eGFR of <30 mL/minute/1.73 m2 all improved their eGFR to values >30 mL/minute/1.73 m2 at discharge.
Table III. Creatinine levels and glomerular filtration rates
| Test | Pre-op | POD #1 | POD #2 | POD #3 | Discharge |
|---|---|---|---|---|---|
| Cr (mg/dL) | 1.1(0.7-2.4) | 1.1(0.7-2.1) | 1.1(0.7-2.4) | 1.1(0.7-2.3) | 1.1(0.6-2.1) |
| eGFR(mL/minute) | 60.7(28.0-137.0) | 63.7(36.6-120.7) | 64.4(30.4-138.0) | 66.4(34.7-137.4) | 66.8(32.1-138.0) |
Fig.
Graph of the in-hospital creatinine levels of the 36 patients who have subsequently been followed for more than 12 months.
There was only 1 patient in the study who was discharged with a Cr ≥2.0 mg/dL. The preoperative Cr was 1.2 mg/dL, increased on POD #2 to 1.9 mg/dL, and 2.1 mg/dL at discharge. This patient's operation was complicated by a known left kidney malignancy requiring subsequent nephrectomy and a 5 mm accessory right lower pole renal artery that was reimplanted into the aortic graft (total suprarenal cross-clamp time was 30 minutes).
Long-term follow-up (>12 months after the original surgery) has been obtained in 36 patients (minimum 12 months, median 34.5 months, maximum of 144 months). Median Cr was 1.1 mg/dL preoperatively and postoperatively. Median preoperative eGFR was 61.2 mL/minute/1.73 m2 and increased to 64.7 mL/minute/1.73 m2 postoperatively. Six patients satisfying this criterion have been lost to follow-up. All of the patients, except 2, have maintained their baseline Cr and eGFR levels. Both of these patients had preoperative Cr levels of 1.5 mg/dL and maintained these levels for 25 and 34 months before their levels increased to 2.1 mg/dL and 2.4 mg/dL, respectively. Nephrology consultation suggests that these changes were due to intrinsic hypertensive nephrosclerosis and were unrelated to renal vein ligation or renal artery stenosis. None of the 36 patients have developed macro or microscopic hematuria and none have reported left flank pain.
Discussion
The complexity of open aortic surgery will continue to increase as endovascular aortic techniques and devices continue to improve. Hence vascular surgeons performing open surgery on the abdominal aorta will more frequently be challenged by short, wide, calcified, or otherwise hostile juxta-renal aortic necks. Temporary division of the LRV has been described as one technique to enhance exposure and control of the juxtarenal aorta and this technique is increasingly being utilized. What remains controversial is whether continuity of the vein needs to be restored since collateral flow is usually preserved through the phrenic, gonadal, adrenal, and lumbar veins provided that these veins are not also divided or absent. Proponents of LRVDAL have described it as a safe adjunct to aortic surgery6 without significant in-hospital deterioration in renal function.7, 8, 9 However, other authors have opposed LRVDAL and reported renal derangements such as an increase in BUN/Cr, decrease in eGFR, venous congestion causing kidney swelling, and sometimes capsular rupture.10, 11, 12, 13, 14
In order to determine whether the vein could be divided and safely ligated, Calligaro et al15 described “test clamping” the renal vein to check stump pressures. They suggested that when the LRV is distended and stump pressure is ≥50 cm water, renal vein ligation and division should not be attempted, rather relying on re-anastomosis or retraction of the vein which can be mobilized more readily by ligating the gonadal vein. More recently, there have been further recommendations that the LRV should be reconstructed after temporary division.16 However, it has been our experience that re-anastomosis is not only unnecessary, but increases both operative time and potential blood loss in an already complex operation. The benefit of LRVDAL on these aspects of the perioperative experience cannot be assessed from this study since we have no valid personal comparison to surgery without this maneuver. However, we believe that the 0% mortality, short cross-clamp times, standard blood loss, and short ICU and hospital stays all attest to excellent exposure afforded by LRDVAL.
Importantly, our series of 56 patients who had LRVDAL failed to demonstrate any significant deterioration in renal function either early or late. Further, although not statistically significant, median discharge eGFR increased slightly in the overall study population and all but 1 of the patients who had renal insufficiency (Cr ≥1.5 mg/dL) experienced a decrease in Cr and an increase in eGFR. This was likely secondary to aggressive perioperative hydration in combination with relatively short suprarenal aortic cross-clamp times. Only 1 patient experienced a clinically significant increase in Cr and decrease in eGFR, but that patient had a malignant tumor in the left kidney and required accessory lower pole arterial re-anastomosis to the graft. These findings are even more convincing when considering that 51 (91%) of the patients also had suprarenal cross-clamp application with vessel loop temporary control of renal artery blood flow in conjunction with LRV division, a combination that is thought to be detrimental to renal function. However, we believe that vessel loop control of the renal arteries is critical to prevent atheroembolism and consider it possible that failure to use this protective method may account for renal failure sometimes reported in other manuscripts dealing with LRVDAL.
Further, unlike previous reports on the subject, we have been able to follow 36 patients for more than a year after LRVDAL to evaluate if there were any long-term effects on renal function. Thirty-four showed no change in renal function and the 2 patients that did develop elevation in Cr and decreasing eGFR both had this occur more than a year after surgery and unrelated to renal vein ligation. Another theoretical long-term risk of LRVDAL is the development of perinephric venous dilation, varicosities, and ultimately hematuria. This has been described in patients who have the so-called “nutcracker syndrome” ostensibly due to compression of the left renal vein against the aorta by the superior mesenteric artery.17, 18 These patients may experience left flank pain with radiation to the left buttock, pelvic pain, and hematuria. Treatment by either stenting techniques or left renal vein transposition have been described.19, 20 However, none of the 36 patients followed long-term have developed any similar symptoms or findings. The only hematuria that our patients experienced appeared to be secondary to trauma from Foley catheter insertion at the time of surgery. We have no explanation for the lack of such venous congestion complications other than to suggest that surgical division immediately next to the inferior vena cava spares collaterals that may be impinged by the superior mesenteric artery in the “nutcracker syndrome” or that patients with this syndrome have some other abnormality of renal venous drainage.
There are potential shortcomings of our review. Only 3 patients had a preoperative eGFR of <30 mL/minute/1.73 m2. Although all 3 patients and the other 5 patients who had eGFR >30 mL/minute/1,73 m2 but a Cr of >1.5 mg/dL all improved their renal function at discharge, it is possible that LRVDAL may not be safe in this subset of renal impaired patients. Also, urinalysis was only performed when hematuria was suspected so long-term microscopic hematuria may have been present in some of the patients.
Conclusion
The LRV can be readily divided adjacent to the inferior vena cava in order to provide improved exposure during complex aortic reconstruction. In our series, suprarenal cross-clamping with temporary cessation of renal blood flow in combination with permanent LRVDAL did not appear to adversely affect early or late renal function. Theoretical disadvantages of renal vein ligation such as flank pain or hematuria also were not noted in this series. Accordingly, we now have a very low threshold to perform LRVDAL whenever improved exposure of the juxtarenal aorta is required.
Author contributions
References
- . Evaluation of the complexity of open abdominal aneurysm repair in the era of endovascular stent grafting. J Vasc Surg. 2006;43:915–920discussion 920
- . Division of the left renal vein as an adjunct to resection of abdominal aortic aneurysm. Am J Surg. 1967;113:763–765
- . Temporal transaction of the left renal vein: a technical aid in aortic surgery. Surgery. 1969;65:32–40
- . Assessing kidney function – measured and estimated glomerular filtration rate. N Engl J Med. 2006;354:2473–2483
- . Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16:31–41
- . Division of the left renal vein: a safe surgical adjunct. Surgery. 1978;83:151–154
- . Left renal vein division in abdominal aortic aneurysm operations (Effect on renal function). Arch Surg. 1985;120:1033–1036
- . Renal artery clamping and left renal vein division during abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg. 2004;27:80–83
- . Effect of left renal vein division during aortic surgery on renal function. Ann R Coll Surg Engl. 2000;82:417–420
- Factors affecting outcomes of open surgical repair of pararenal aortic aneurysms: a 10-year experience. J Vasc Surg. 2006;43:921–927discussion 927-8
- . The risk of ligation of the left renal vein in resection of the abdominal aortic aneurysm. Surg Gynecol Obstet. 1991;173:33–36
- . Does division of the left renal vein during aortic surgery adversely affect renal function?. Ann Vasc Surg. 1991;5:74–79
- . Technical and operative factors in infrarenal aortic reconstruction and their effect on the glomerular filtration rate in the immediate postoperative period and 6 months later. Eur J Vasc Surg. 1990;4:239–245
- . Renal complications to left renal vein ligation in abdominal aortic surgery. J Cardiovasc Surg (Torino). 1984;25:432–436
- . Division of the left renal vein during aortic surgery. Am J Surg. 1990;160:192–196
- . Glomerular filtration rate after left renal vein division and reconstruction during infrarenal aortic aneurysm repair. J Vasc Surg. 2007;45:481–486
- . Anatomical and surgical aspects in the operative management of varicocele. Urol Cutaneous Rev. 1950;54:257–262
- . [“Nutcracker” phenomenon of the renal vein and venous pathology of the left kidney.] [Article in Dutch]. J Belge Radiol. 1972;55:507–511
- . The nutcracker syndrome: its role in the pelvic venous disorders. J Vasc Surg. 2001;34:812–819
- . Transposition of the left renal vein for treatment of the nutcracker phenomenon: long-term follow-up. Urology. 2002;59:354–357
Competition of interest: none.
PII: S0741-5214(09)00893-3
doi:10.1016/j.jvs.2009.04.041
© 2009 Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
