Abstract

Herein, we report a rare instance of paraplegia following percutaneous nephrolithotomy under thoracic epidural anesthesia in a conscious patient. The possible factors include low body mass index, chronic renal failure, and multiple “in and out” needle passes during the procedure. Direct cord trauma with epidural bloody collection resulted in the neurological damage. However, the patient did not show signs of spinal cord trauma during the needle injury, possibly due to a prior accidental subarachnoid block. This not only delayed the recognition of the disorder, but also delayed treatment, consequently resulting in permanent paraplegia. Careful monitoring after epidural blocking should be undertaken in order to allow the early detection of mismanagement and limit the extent of neurologic injury.

Keywords

anesthesia, epidural; paraplegia; nephrolithotomy; percutaneous;

1. Introduction

The incidence of permanent neurological complications following thoracic epidural catheterization is low, with an incidence < 0.02%.2, 3, 5, 7 The neurological complications include postoperative radicular pain, peripheral nerve lesions, and permanent sensory and motor loss. The causes are diverse, including epidural hematoma, epidural abscess, direct cord trauma, spinal infarction, and neurotoxicity by accidental subarachnoid injection or chemical contamination.2, 7 Herein, we report the outcome of a rare case of spinal cord contusion following thoracic epidural anesthesia during percutaneous nephrolithotomy (PCNL) and describe possible methods that could be used to prevent such a debilitating complications.

2. Case report

A 40-year-old female weighing 35 kg and standing 145 cm with a body mass index (BMI) 16.6 kg/m² was admitted for the management of bilateral staghorn calculi (Fig. 1) with infected hydronephrosis, chronic renal failure (blood urea: 99 mg/dL; serum creatinine: 5.1 mg/dL), and worsening coagulation parameters (bleeding time: 2 minutes and 30 seconds; PT: 22 seconds; international normalized ratio (INR): 1.57). She was not on any antiplatelet drugs. Three months later, her renal function had stabilized, demonstrating a creatinine level of 2.2 mg/dL after bilateral percutaneous nephrostomy tube placement, and her coagulation profile was normalized after the transfusion of 4 units of fresh frozen plasma and vitamin K injection for 7 days. She initially underwent left PCNL under general anesthesia (GA), but significant residual fragments remained. She was schedule to undergo a second left PCNL again 1 week later. Because the patient had a respiratory tract infection at the juncture, the procedure was planned to be performed under epidural anesthesia (EA). Her preoperative vital parameters were within normal limits (pulse rate: 70 beats/minute; blood pressure: 112/74 mmHg). Throughout the procedure she was monitored using a 5-lead electrocardiogram, pulse oximetry, and noninvasive blood pressure recordings. Prior to EA, the skin where the epidural needle was to enter was infiltrated with 2 mL of 2% lignocaine with aseptic precautions. An epidural block was administered between the T12–L1 interspace using an 18-gauge Tuohy needle. Four passes were needed to reach the epidural space due to the loss of resistance to the air technique. A single-end port catheter was placed 4 cm inside the space and at the mark of 8 cm at the skin. After confirming the position by negative blood and cerebrospinal fluid (CSF) aspiration, a test dose of 3 mL of 2% lignocaine with adrenaline (1: 2,00,000) was administered. No changes in the heart rate or blood pressure were noted. There was no motor blockage. The epidural catheter was secured at the midline. The patient never complained of paresthesia during needle placement or injection. Ten minutes after the test dose, an epidural top-up dose of 10 mL of 0.5% bupivacaine and 10 mL of 2% lignocaine was administered in aliquots. Blocking up to T8 was achieved, and the segmental level of anesthesia was assessed using an alcohol swab.

Download full-size image

Fig. 1. Bilateral staghorn calculi.

The surgery lasted for 2 hours, during which the patient’s pulse and blood pressure remained stable. The patient remained conscious throughout the procedure. At the end of procedure, sensory loss had regressed to the level of the L1 segment. The patient was observed in the postoperative room for 1 hour, then returned to the ward. During the evening rounds 8 hours later, the patient complained of bilateral complete sensory and motor loss in the lower limbs. Neurological assessment revealed complete sensory loss and paraplegia below the level of the L1 segment. The epidural catheter was removed after negative aspiration for blood and CSF, and the length of the catheter was 4 cm inside, thus ruling out spontaneous migration of the catheter. The postoperative coagulation profile was within normal limits.

The patient also reported the complete loss of bowel and bladder sensations. Sagittal and axial T2 magnetic resonance imaging (MRI) were performed, demonstrating anterior epidural collection at the level corresponding to T11–T12 and cord contusion in the dorsolumbar spinal cord extending from T9–L1 (Fig. 2). Following neurosurgical consultation, a loading dose of 30 mg/kg methylprednisolone was administered over 15 minutes, followed by 5.4 mg/kg/hour intravenous infusion over in the next 23 hours. The neurological deficits were evaluated daily. Neurological examination after 1 week revealed the persistence of complete sensory loss with paraplegia below L1 level. The patient did not show any signs of sensory or motor improvements even 2 months after surgery.

Download full-size image

Fig. 2. Sagittal T2-weighted MRI performed 8 hours after EA. The solid arrow indicates anterior epidural collection at the D11/D12 level, and the dashed arrow indicates cord contusion in the dorsolumbar spinal cord extending from D9–L1.

3. Discussion

There is an increasing trend of performing PCNL under regional anesthesia. Meharabi evaluated 160 consecutive patients who underwent PCNL under spinal anesthesia in the prone position and concluded that spinal anesthesia is safe and effective for performing PCNL and is a good alternative to general anesthesia in adult patients.¹¹ In another study, Kuzgunbay noted that combination spinal-regional anesthesia is a feasible technique for performing PCNL operations because the efficacy and safety are not inferior to general anesthesia.¹⁰ However, this techniques carries the risk of neurologic injury, and the overall incidence of complications following thoracic epidural catheterization has been reported as 3.1% and the predicted maximum risk for permanent neurologic complications is 0.07%.⁵ Kao and others reported a case of paraplegia following thoracic epidural catheterization for abdominal surgery in an elderly patient, and the cause of the misfortune was intracord catheterisation followed by local anesthetic injection.⁸

Spinal cord contusion leading to permanent paraplegia is an unusual and rare complication that occurs as a consequence of inadvertent EA. The possible risk factors for neurologic injury in our case were low BMI, chronic renal failure, and multiple needle passes when performing the procedure. The distance from the skin to the ligamentum flavum is usually 4 cm, with the depth in most (80%) patients reported as between 3.5–6 cm.⁴ Low BMI, as in our case, might be predictive of a shorter distance between the skin and the epidural space. Proper control of needle insertion is imperative, and ultrasound guidance can be used to avoid inadvertent spinal cord injury.⁶ Some studies have suggested that performing multiple attempts in order to administer a technically difficult block is also a risk factor.⁹ Decreased renal function can lead to abnormal platelet function, resulting in epidural bleeding or hematoma with eventual cord compression. However, in our case, the coagulation profile, including bleeding time both before and after the operation, was within the normal range (Fig. 3).

Download full-size image

Fig. 3. Axial T2-weighted MRI image. Arrow indicates a hyperintense signal intensity suggestive of a cord contusion.

Previously, all case reports of spinal cord injury following regional block in conscious patients mentioned that the patient complained of paresthesia as a result of injury to the spinal cord.¹ In our case, direct traumatic injury to the cord was the cause of paraplegia, as revealed by the MRI findings. However, it was unusual that our patient never complained of localized radiating pain or paresthesia during the procedure as the patient was conscious. The local anesthetic procedure could be one of the reasons that contributed to such an injury. Because the patient was short in stature, it is possible that the local anesthetic, which was delivered via subcutaneous injection, was accidently injected into the subarachnoid space before epidural needle insertion, thereby leading to sensory loss. Therefore, our patient could not feel any painful sensation or paresthesia even during the traumatic misplacement of the epidural needle. Thus, we emphasize that spinal cord injury can occur without any warning signs during epidural needle insertion in a patient with low BMI.

Other possible contributing factors include intracord catheterization and intraneural injection of local anesthetics, which could lead to permanent cord damage. The delayed response from the patient, as well as the lack of a regular postoperative neurological examination during the postoperative period, might also have contributed to the permanent neurological injury. Although systemic steroids were administered, the patient did not recover, at least in part due to the delay in administration.

Thus, EA should be carefully performed and adhere to the rules of the technique. Clinicians should be aware of rare but devastating neurological complications that can occur in conscious patients without any warning signs. This case undermines some of the unusual matters that can arise during the use of epidural anesthesia for PCNL procedures, especially in high-risk patients. Postoperative examination to guard against neurological recovery should be routinely performed for the early detection of complications and to limit the extent of injuries, should they occur.

Financial support

None.