UCLA David Geffen School of Medicine

Chief Complaint:

Back pain on post op day 1 status post right total hip arthroplasty. She had a history infected right native hip, status post right hip resection arthroplasty with spacer placement procedure 4 months ago (first stage), now in need of total hip arthroplasty (second stage)

Hx of infected right native hip, status post right hip resection arthroplasty with spacer placement procedure 4 months ago (first stage), now in need of total hip arthroplasty replacement (second stage).

History of present illness

A 36-year-old female with end-stage renal disease from hypertension, previously had a systemic methicillin-resistant staphylococcus aureus (MRSA) infection due to an infected hemodialysis catheter. She eventually developed a right hip MRSA infection that destroyed her joint. She underwent a two-stage approach. Stage I was a resection arthroplasty with a Girdlestone procedure. This included placing an antibiotic-loaded spacer into her acetabulum and down her femoral shaft. She required treatment with organism-specific IV antibiotics for a minimum of 8 weeks. After this, she would be considered for a total hip arthroplasty 3 to 4 months after the initial procedure. If she were unable to clear her infection, she would be left with a Girdlestone, a right native hip resection with spacer placement.

She was treated postoperatively with daptomycin for approximately 8 weeks and another 8 weeks with doxycycline. She was followed up by the infectious disease consultant, her ESR and CRP was improving and she was scheduled to undergo a total hip replacement. The patient had been living in a skilled nursing facility since her 1st stage surgery. She was having pain with movements and with touchdown toe bearing on her right leg. She had been working with physical therapy and had been trying to use a walker. She had no major medical illnesses since her previous surgery. She was on hemodialysis 3 times a week, and her last dialysis was a day before surgery.

As far as the patient's chronic medical conditions, in addition to hypertension, dialysis-dependent ESRD, and the treated MSRA infection. She specifically denied a history of heart attack, coronary artery disease, congestive heart failure, stroke, easy bruising or diabetes. She tolerated her previous surgery with no cardiovascular complications. As far as her functional status, she had previously used a walker to ambulate short distances, but was wheelchair-bound.

Past Medical History

1. End-stage renal disease from hypertension, on hemodialysis
2. History of a systemic MRSA infection. She has been on chronic renal dialysis, most recently via a left upper extremity AV fistula. She previously had line epsis associated with the right subclavian dialysis catheter. This was believed to be a source of infection in her right hip. She was noted to develop a right septic hip that was treated 2 years ago with operative irrigation and debridement. She was treated with vancomycin for 8 weeks after this and then was noted to have a persistent infection; an aspiration confirmed the findings of infection. She had been treated with oral Zyvox, which she continues on at this time. She has had progressive hip and back pain and decreased mobility because of the right hip. Her radiographs showed severe degenerative changes and obliteration of her femoral head. She is continuing to have severe back and hip pain and her mobility is limited. She has been using a wheelchair for the past 3 months. She is requiring hydromophone (Dilaudid) injections for continued pain relief.
3. Asthma
4. Anemia
5. Hypertension
6. Chronic pain
7. Tobacco abuse

Past surgical history:

1. Status post right hip resection 4 months ago under regional anesthesia (CSE).
2. Status post right hip irrigation and debridement 2 years ago.
3. Status post AV fistula placement.
4. Status post tubal ligation.

She denies having trouble with anesthesia, bleeding problems, or blood clots.

Medications

Nephrovite 1 tab daily
Vit B complex w/vit B12 daily
Vasotec 20 mg bid
Epogen w/HD
Clonidine 0.1mg every 8 hours
ASA EC 81 mg daily
Doxazosin 8 mg BID
Furosemide 40 mg daily
Nifedipine ER 60 mg BID
Renagel 4000 mg TID w/meal
Nepro 1 can daily
Pro-Stat 30 ml BID
Colace 200 mg BID
Doxycyline 100 mg BID - stopped 1/14/08
Klonopin 1 mg BID
Albuterol 2.5 mg HHN q6 hr PRN
Ipratropium 0.5 mg HHN q6 hr PRN
Vicodin 5/500 mg q 4 hr 2 tabs
Dilaudid 1 mg IM q6 hrs PRN breakthrough pain
Oxycodone 5 or 10 mg q 3hr for breakthrough pain
Compazine 10 mg q8hr prn n/v

Allergies: No known drug allergies.

Family History:

Non-contributory

Social history:

The patient is divorced. She has 7 children. She is living in a nursing facility at this time. She is to work as a caregiver but is currently not working. She smokes half a pack per day for about 1 year. She does not drink alcohol or use recreational drugs

Review of Symptoms:

Not significant.

Physical exam:

Vital Signs: Blood pressure 142/95, pulse 78 and regular, temperature 97.6, oxygen saturation 98% on room air, weight 180 pounds and height 5'7". General: The patient is in no acute distress and looks chronically ill She is alert and oriented x3. HEENT: Eyes: Pupils are equally round and reactive to light. The sclerae are anicteric. Ears, Nose, Mouth, and Throat: Oropharynx is clear. There is no nasal discharge. Neck: Supple. There is no thyromegaly. Respiratory: Lungs are clear to auscultation bilaterally. There are no retractions. Cardiovascular: regular rate and rhythm; no lower extremity edema. There is a 2/6 systolic ejection murmur heard loudest at the apex. Abdomen: Soft, nontender and nondistended. No masses or organomegaly are appreciated. The abdomen is obese. Skin: No obvious rash. Skin is smooth to palpation. Extremities: No edema. She does have a dialysis shunt in her left upper extremity. Lymph Nodes: There is no obvious cervical, supraclavicular, or infraclavicular lymphadenopathy.

Laboratory Data:

White blood cell count 5.8, Hemoglobin 11.1, Hematocrit 44.2, Platelets 274, sedimentation rate is 9, PTT 26.1 and INR 1.1. Potassium 5.3, BUN 61, and creatinine 10.8. Urinalysis has greater than 300 protein, trace blood, trace reducing substance and 5 to 7 white blood cells.

Radiological Data:

Right Hip X-rays: Severe degenerative changes and obliteration of her femoral head.

EKG:

Normal sinus rhythm at a rate of 69. No Q waves or ischemic changes.

• End-stage renal disease from hypertension, on hemodialysis who previously had a methicillin-resistant staphylococcus aureus (MRSA) systemic infection related to her hemodialysis catheter and eventually developed a right hip MRSA infection that destroyed her joint.
• Hypertension
• MRSA
• Asthma
• Chronic pain
• Limited activity
• Anemia

We discussed with the patient the following anesthetic options: regional anesthesia with MAC, general anesthesia with regional anesthesia, general anesthesia alone. The patient preferred to proceed with the same anesthetic technique she had received with her prior surgery: regional anesthesia with MAC. We planned on using standard ASA monitors during induction, maintenance, and emergence from anesthesia.

We placed the patient in the sitting position on the operating room table, placed standard ASA monitors and 6 L/min oxygen by face mask on the patient. Invasive monitoring of arterial blood pressure was discussed by the anesthesia team, but an arterial line was not placed in the upper extremities in order to preserve vessels for potential future AV fistulas. The option of a lower extremity dorsalis pedis arterial line was felt to be suboptimal secondary to infectious risk and poor access to it. Therefore, monitoring of blood pressure during the case was via a noninvasive sphygmomanometer which was cycled every 1- 2 minutes during critical periods, otherwise every 3 minutes.

The thoracolumbar region was prepped and draped sterilely for the combined spinal/epidural. Local was applied to the skin at the L3/4 level with 1 ml lidocaine 1%. The 17 gauge Tuohy needle was placed atraumatically using a midline approach with the loss of resistance technique with saline/air mixture. The epidural space was located with one attempt. Then a 27G Whitacre spinal needle was advanced through the Tuohy needle until CSF was visualized in the spinal needle. No paresthesias and no blood were obtained from either the Tuohy or the spinal needle. 1.6ml hyperbaric bupivacaine 0.75%, with 20mcg fentanyl were administered intrathecally after good CSF flow with syringe aspiration. The spinal needle was removed, and an 18 gauge wire-reinforced end-holed epidural catheter was advanced 4cm into the epidural space via the Tuohy needle. The Tuohy needle was removed from the patient’s back, and the epidural catheter was taped to the patient’s back with sterile tegaderm.

We used a hyperbaric local anesthetic for the spinal in order to be able to better control the level of the dermatomal spread with changing the patient’s position on the OR table. The hip surgery itself only requires dermatomal blockade below T11. However, if the case were prolonged, and we wished to take advantage of the surgical anesthesia provided by the spinal for the longest period of time without excessive sympathectomy and hypotension, we optimally aim to achieve a higher level of sensorimotor blockade at a higher thoracic level, such as T6. The optimization of the duration of the level of sensory blockade with a hyperbaric spinal has not been specifically addressed in the literature with all patient parameters being similar. However, literature data suggests that adding epinephrine to hyperbaric spinal bupivacaine does not predictably prolong its duration. However, adding epinephrine to most other spinal local anesthetics, including isobaric bupivacaine, does prolong the duration of action.

Isobaric spinal medication could also be a good option, if the patient will be sitting up for a prolonged period of time during the epidural catheter advancement after the spinal. Hypobaric spinal medication is also a good alternative if the patient will remain in the lateral decubitus position with the operative side up, such as with a fractured hip.

Sensory blockade to light touch and loss of cold sensation from the spinal was confirmed below the T8 dermatome. After initial surgical incision was made, the patient became hypotensive and required a phenylephrine infusion to support her blood pressure, followed by 2 units PRBC. The patient was able to communicate throughout this period of hemodynamic instability, and then was hemodynamically stable for the remainder of the case. The target range of acceptable blood pressure varies from patient to patient, depending upon comorbidities as well as starting values and lability of blood pressure. One of the greatest advantages of regional anesthesia is the ability to monitor serial neurologic exams as a means of assessing cerebral perfusion during periods of critical hypotension. With adequate patient communication and responses, we were able to gauge her ability to tolerate the hemodynamic swings during the case. During the five-hour intraoperative period, her systolic blood pressure ranged from 80-148, and her diastolic blood pressure ranged from 35-80. Her baseline preoperative blood pressure was 140/75. However, it seems reasonable that the maintenance of cardiac stroke volume and inotropy with regional anesthesia may be an underlying mechanism to explain the preservation of cerebral perfusion during periods of hypotension. Of course, eventual preload replacement is essential in order to maintain ventricular filling pressures.

We opted to use the same regional anesthetic, a CSE that had worked well without complications for this patient in the past. This technique works well daily for the majority of our patients undergoing hip replacement surgery. The other anesthesia options include general anesthesia, spinal alone (If procedure is not prolong), or epidural alone.

This is a controversial topic, with many varying opinions in the literature. Some of the advantages of the CSE include solid spinal block for surgery, the higher epidural catheter success rate than when an epidural only is placed, as well as the ability to offer the patient postoperative analgesia, compared to the finite duration of a spinal alone. The higher reported catheter success rate with CSE’s has an intuitive anatomical foundation.

Anatomically speaking, the cerebrospinal fluid obtained from the spinal needle which is passed through the Tuohy needle is a confirmatory sign that the Tuohy needle should have passed through the epidural space. Intraoperative use of epidural anesthesia, with any of these three techniques, has been shown to decrease intraoperative blood loss as well as decrease the risk of perioperative thromboembolism and fatal pulmonary embolism and improve bowel function post op. The decreased intraoperative blood loss has been demonstrated for a wide range of surgical procedures is postulated to be due to absence of venous congestion and high hydrostatic pressure (as seen with positive pressure ventilation under general anesthesia).

In this patient who had been exposed to narcotics preoperatively, the epidural provided an alternative method to IV narcotics only for postoperative analgesia. The patient was placed on PCEA bupivacaine 0.125% with fentanyl 5mcg/ml (8 cc/hour continuous and 3cc q12 mins PCEA) and resumed her oral regimen. IV PCA Hydromorphone was added due to her NPO status, high opioid tolerance and prevent potential withdrawal. (Fentanyl is a better choice because of its lack of active metabolites and thus potential for systemic toxicity in a patient with ESRD.) Since this patient had been on chronic high-dose narcotics, she was very likely to go into withdrawal if her home dose of baseline narcotic had not been maintained as well as supplemented with additional analgesia for surgical pain.

The patient was transported to the ICU at the end of the case, hemodynamically stable, awake, and with pain well-controlled via the PCEA infusion of bupivacaine 0.125% with fentanyl 5 mcg/ml. and resumed on home meds (including Vicodin(R) and oxycodone). The patient received a routine, but reduced dose of warfarin 0.5 mg once for deep vein thrombosis prophylaxis. The patient still complained of pain, so IV PCA hydromorphone was added. The patient was seen in hospital by the acute pain service attending for epidural catheter management, when she began to complain of back pain. The onset of progressively severe low back pain occurred ten hours after surgery with no motor or sensory deficits. Pain radiated from the back towards the umbilicus. The pain was so severe that the patient insisted on the removal of the epidural catheter even before performing MRI. Physical examination revealed nonspecific tenderness on multiple levels of the thoracolumbar region. The site of the epidural catheter was clean and dry with no signs of erythema or induration. Induration is one of the most important signs of catheter infection.

• Epidural hematoma
• Disc disease
• Epidural abscess
• Worsening of her baseline back pain
• Back injury

An MRI was requested, and the epidural catheter had to be removed per MRI protocol and epidural catheter manufacture recommendation in order to safely obtain the scan. The catheter’ wire-reinforced portion may potentially move in the magnetic field of the MRI. The catheter was removed, and an MRI was obtained. According to the radiologist, the MRI demonstrated a lumbar 3-4 level epidural hematoma, which was causing posterior mass-effect and indentation on the ventral nerve roots, measuring approximately 9 mm AP x 9 mm TR x 19 mm AP. A small rim of T2 hypointensity was seen around this lesion, suggesting hemorrhagic products. Axial T2-weighted images demonstrated moderate to severe central canal stenosis with deformation of the thecal sac and probable compression on the descending nerve roots. There was effacement of the lateral recesses at this level with probable impingement on the traversing L3 nerve root.

Spine surgery was consulted. Since there were no neurological deficits, the patient was closely monitored during the rest of her hospitalization. She received a neurological exam hourly. This included lower extremity sensorimotor function, symptoms of back pain exacerbation, as well as assessment for bowel or bladder incontinence. A follow-up MRI three days afterward showed a significant interval decrease in size of the lumbar epidural hematoma. The patient was discharged on post-operative day seven without any neurological compromise from the hematoma.

Initial MRI
Follow up MRI

Isolated back pain usually signals a hematoma in the posterior epidural space which is impinging on nerve fibers to the posterior structures. Back pain along with lower extremity pain and/or sensory or motor deficit heralds a hematoma in the posterolateral epidural space which is also impinging on the nerve roots. If the hematoma blocks the entire spinal canal, it will cause cauda equina syndrome with urinary or fecal incontinence and weakness of lower extremities which is associated with complete obliteration of the thecal sac.

The wire-reinforced epidural catheter is usually not MRI compatible, since the metal component is ferromagnetic and drawn toward the magnet. The patient’s epidural catheter had to be removed prior to her having the MRI scan.

Why some patients may be more prone to epidural hematoma than the normal population?

End-stage renal disease and uremia are risk factors for epidural hematoma. Moreover, this patient demonstrates the systemic ramifications of advanced renal disease in multi-organ dysfunction, including the coagulation system. Even though the patient may have a normal platelet count, uremia causes a qualitative platelet dysfunction. In fact, advanced renal disease disrupts normal platelet activation, aggregation, and adherence. The etiology of altered hemostasis is multifactorial, but includes a release in platelet-inhibitory factors as well as a disturbance in the platelet-vessel wall.

Early studies had made the observation that uremic platelets in normal plasma function normally. Consequently, uremic plasma mixed with normal platelets triggered platelet dysfunction. Thus, the hypothesis was made that platelet-inhibitory factors in the plasma of uremic patients led to impaired platelet function. However, in vitro studies could not replicate this effect with the known uremic toxins until the discovery of guanidinosuccinic acid. The formation of guanidinosuccinic acid occurs in the urea cycle when excess urea exerts an inhibitory effect on the cycle’s own enzymes. Therefore, the accumulation of L-arginine, an intermediate in the urea cycle, gets converted to guanidinosuccinic acid in an alternate pathway for ammonia detoxification. Interestingly, L-arginine is also a substrate for nitric oxide production. Excess nitric oxide production by the vascular endothelium and uremic platelets is implicated as the cause of bleeding by inhibiting platelet aggregation. Guanidinosuccinic acid is thought to be a precursor to nitric oxide production and may actually mimic the effects of nitric oxide as well.

The goal of treatment prior to surgery is to reduce platelet dysfunction in patients who are about to undergo an invasive procedure. In addition to the basic labs in the preoperative workup, the bleeding time is a nonspecific test that can be used to evaluate platelet and vascular wall function. Although uremic bleeding is associated with a prolonged bleeding time, it is controversial whether the test is a reliable predictor of clinical bleeding, such as an epidural hematoma. Unfortunately, there have been no large prospective studies that correlate prolonged bleeding time to risk of bleeding secondary to a procedure. The bleeding time can be used to evaluate the effectiveness of therapy. One such treatment is desmopressin (DDAVP) which is an analog of antidiuretic hormone. Desmopressin’s mechanism of action occurs at the endothelium where the drug increases the secretion of von Willebrand factor. Given at least one hour prior to any invasive procedure at a dose of 0.3 micrograms/kilogram intravenously, its duration of action persists for four to twenty-four hours. The advantage of desmopressin is that it can be used in an acute setting where the first dose can lower the bleeding time in half of the patients. Routine dialysis preoperatively has reduced the amount of uremic bleeding during surgery. Although dialysis has been shown to improve platelet function by eliminating circulating toxins, it may not effectively eliminate the risk of an epidural hematoma. Dialysis has been shown to partially reduce bleeding time in only two-thirds of uremic patients. Moreover, there have been case reports of spontaneous epidural hematomas during dialysis. In conclusion, even though these treatments can reduce the frequency of uremic bleeding, they won’t reliably normalize the risk of epidural hematoma equal to the general population.

End-stage renal disease patients under neuraxial anesthesia require close monitoring for any signs of spinal cord compression. The monitoring that was provided to the patient by the nursing staff and supervised by the acute pain service, provides the template for future protocols that may be implemented for other patients that have developed signs/symptoms of epidural hematoma. This protocol should include hourly neurological examination for 12 hours followed by exams every 2-4 hours for 24 hours consisting of lower extremity sensorimotor examination, assessment of bowel/bladder function, pre-anal pinprick sensory change (which may be the first sign of cauda equina syndrome), or exacerbation of back pain. Even though the outcome may be catastrophic with an epidural hematoma, the benefits of regional anesthesia greatly outweigh the risks for orthopedic procedures such as joint replacement surgery. The risk-to-benefit ratio must be formulated individually for each patient based on comorbidities as well as the surgical procedure.

The American Society of Regional Anesthesia (ASRA) has published consensus guidelines for the safe placement of neuraxial blocks. These guidelines [http://www.asra.com/consensus-statements/RAPM-Anticoagulation.pdf] include accepted periods for withholding certain anticoagulants prior to neuraxial techniques, as well as acceptable limits on coagulation labs and platelet counts. In this patient, the ASRA guidelines had been followed. Desmopressin had also been used to improve platelet function. She also received aggressive neurologic monitoring and examinations postoperatively for early detection.

The ASRA guidelines aim to prevent the risk of complications such as epidural hematoma after neuraxial blockade. However, there is no guarantee against such an event. There have been documented case reports of spontaneous epidural or spinal hematomas in patients who have not undergone any invasive neuraxial procedure. Early detection and neurological consultation and decompression within 8 hours are critical to reversal of neurologic deficit.

In the absence of any focal neurological deficits, conservative therapy and observation are warranted. However, if the patient exhibits any symptoms of progressively severe back pain with motor or sensory deficits, then a MRI/CT scan and emergency surgery should be performed expediently in order to reduce poor neurological outcome from the epidural hematoma.