To the Editor:-We report a case in which computed tomography (CT) scan-guided superior hypogastric block for management of malignant perineal pain was complicated by sensory impairment. A 57-yr-old woman with carcinoma of cervix, previously treated by hysterectomy and radiotherapy, presented with perineal pain. The pain had been present for 6 months, and was described as burning, heavy, and disturbing to her sleep at night. She had to rely on transdermal fentanyl, 25 micro gram/h with rescue oral morphine up to 70–80 mg/day for pain relief. Her average pain score was 5–7 of 10 on visual analogue scale (VAS), but she could not tolerate a higher dose of opioid because of sedation and troublesome nausea and vomiting. She had a severely kyphoscoliotic spine caused by a previous tuberculous infection.

Diagnostic superior hypogastric plexus block was attempted with the two-needle posterior approach under biplanar fluoroscopic guidance. Because most of the lower lumbar vertebrae were collapsed, it was almost impossible to define the exact anatomy. However, 8 ml of bupivicaine, 0.25%, was injected bilaterally over the anterior aspect of L5-S1 vertebral body. It resulted in complete pain relief for 7–8 h after the block, but it also was associated with partial motor paralysis of her right lower limb and sensory deficit for 2–3 h. The difficulty experienced during the diagnostic block prompted us to plan the neurolytic block under CT scan guidance.

Computed tomography scan was performed with the patient lying in the supine position, aided by pillows to prop up her upper back because of her severe kyphotic deformity. A General Electric (Milwaukee, WI) Hispeed Advantage Plus Spiral CT scanner was used. From lateral and frontal scout images, 5-mm thick contiguous axial scans were obtained from L4 to S1. A point 5 mm caudal to the aortic bifurcation, at S1, was identified. By means of a superimposed grid on the axial image as seen at the CT console monitor and by using corresponding laser light beams at the CT scanner gantry, the skin was surface marked, cleaned, and draped. Without gantry angulation, a 22-gauge Chiba needle was inserted. The final position of the needle tip was located just caudal to the aortic bifurcation, medial to the proximal right common iliac artery. A test dose of 2.5 ml of nonionic contrast medium (Omnipaque 300) confirmed opacification of the retroperitoneal space adjacent to the proximal common iliac vessels (Figure 1). Eight milliliters of phenol, 10%, was then injected followed by 1 ml of saline during withdrawal of the needle. The whole procedure took 1 h.

Figure 1. Axial computed tomography of the upper sacrum shows the position of the needle tip located just posteromedial to the right common iliac vessels, which have been faintly opacified by intravenously administered contrast agent. The retroperitoneal space is delineated by a test dose of Omnipaque 300 (arrowheads). The S1 foramina are arrowed (arrows). Although bilateral spread of Omnipaque 300 is shown, it is more marked on the right side. (Reference image at the bottom right corner shows location of the axial section on the frontal scout image.)

Figure 1. Axial computed tomography of the upper sacrum shows the position of the needle tip located just posteromedial to the right common iliac vessels, which have been faintly opacified by intravenously administered contrast agent. The retroperitoneal space is delineated by a test dose of Omnipaque 300 (arrowheads). The S1 foramina are arrowed (arrows). Although bilateral spread of Omnipaque 300 is shown, it is more marked on the right side. (Reference image at the bottom right corner shows location of the axial section on the frontal scout image.)

Close modal

Two weeks after neurolytic block, she was maintained on oral morphine, 60 mg/day with a VAS score of 1 or 2. However, she was troubled by persistent right lower limb parasthesia after the block. Physical examination revealed decreased sensation to light touch and temperature over the right L4-S1 dermatomes, although there was no motor impairment. The parasthesia was helped by transcutaneous electrical nerve stimulation (TENS) therapy. She also complained of mild abdominal pain in the first 2 days after the neurolytic block, which subsided spontaneously. There was no sign of infection, peritonitis, or urinary incontinence.

Efficacy of the superior hypogastric block in patients with pelvic malignancy and visceral pain had been demonstrated by Plancarte, et al. (1990) and Oscar, et al. (1993), and is further confirmed in our case. In their series, no complications resulting from the block were recorded. [1,2] Although not described in detail, it was presumed that their patients had fairly normal anatomy of the lumbosacral spine. In this case, the severe kyphoscoliotic lumbosacral junction deformity and the semirecumbent position of the patient may have contributed to the somatic nerve damage, possibly by spreading the neurolytic solution posteriorly to involve L5 nerve rami and caudally to damage the S1 nerve rami. The asymmetric collapse and deformity of the lumbar spine probably resulted in spread of more neurolytic solution to right side (Figure 1). With the experience in our case, we suggest that caution should be exercised in patients with severely deformed lumbosacral spines, and perhaps a smaller volume of neurolytic agent should be used.

W. S. Chan, M.B.Ch.B., F.A.N.Z.C.A., F.H.K.C.A., F.H.K.A.H.

Senior Medical Officer, Department of Anesthesiology

Wilfred C. G. Peh, M.B.B.S., D.M.R.D., F.R.C.R., F.H.K.C.R., F.H.K.A.M., F.A.M.S.

Associate Professor, Department of Diagnostic Radiology

K. F. J. Ng

S. L. Tsui, Professor J.C.S. Yang, M.D., Dip Am Board, F.F.A.R.C.S.(I), F.H.K.A.M.

Department of Anesthesiology

FZ Block; Queen Mary Hospital

102 Pokfulan Road; Hong Kong

(Accepted for publication March 26, 1997.)

1.
de Leon-Casasola OA, Kent E, Lema MJ: Neurolytic superior hypogastric block for chronic pain associated with cancer. Pain 1993; 54:145-51.
2.
Plancarte R, Amescua C, Patt RB, Aldrete JA: Superior hypogastric block for pelvic cancer pain. Anesthesiology 1990; 73:236-9.