POSTDURAL puncture headache (PDPH) is a complication of spinal anesthesia and unintentional dural puncture during attempted epidural anesthesia. We describe a case of PDPH after unsuccessful epidural placement, followed by successful combined spinal–epidural placement. Conventional treatment of PDPH failed. Successful treatment was obtained by administering intravenous cosyntropin, a synthetic form of adrenocorticotropic hormone (ACTH).
A healthy 19-yr-old term parturient (gravida II, para 0) was admitted to the hospital in active labor after spontaneous rupture of membranes. She requested epidural analgesia. Preassessment showed a healthy individual (weight, 80 kg; height, 168 cm; American Society of Anesthesiologists physical status, 2) with an uneventful pregnancy. The patient denied any significant medical history, specifically headaches or chronic back problems.
Epidural catheter placement was attempted using an 18-gauge Weiss epidural needle inserted perpendicular to the dural fibers at L3–L4 without success. Two attempts were made at L3–L4 using a loss of resistance to air technique. There was no loss of resistance or evidence of unintentional dural puncture. A successful combined spinal–epidural was then placed at L4–L5 using an 18-gauge Weiss epidural needle with the bevel inserted perpendicular to the dural fibers and using the loss of resistance to air technique. A Sprotte 24-gauge spinal needle was then placed through the Weiss needle, and free-flowing cerebrospinal fluid was obtained. Satisfactory analgesia was achieved with intrathecally administered bupivacaine 2.5 mg and fentanyl 25 μg. An epidural closed-end, multiorifice, nonstyleted, 19-gauge catheter was inserted 5–6 cm through the Weiss needle and left in place. A test dose of 3 ml xylocaine 1.5% with epinephrine 1:200,000 was administered through the catheter to rule out intrathecal or intravascular catheter placement. Analgesia was maintained with a continuous epidural infusion of bupivacaine 0.125% and fentanyl 2.5 μg/ml at 10 ml/h. The patient remained comfortable throughout labor and had a spontaneous vaginal delivery 4 h later. The epidural catheter was removed immediately after delivery.
Approximately 12 h after delivery, the patient began to complain of an occipital headache that radiated to her shoulders. An anesthesiologist was consulted for evaluation of possible PDPH. The headache worsened when the patient was in the upright position. The patient reported associated nausea. She denied vomiting, photophobia, tinnitus, or symptoms of cranial nerve involvement. Physical examination and vital signs were within normal limits. A diagnosis of PDPH was made. She was treated with Ringer’s lactate 125 ml/h, acetaminophen with codeine as needed for pain, and caffeine sodium benzoate 500 mg in 1 l intravenous fluids per day. The patient reported mild relief of symptoms. She refused further treatment for her headache was discharged home with a prescription for analgesics and instructions to return should her headache worsen.
The patient returned on day 2 complaining of severe headache. Neurologic/physical examination was unchanged from the previous day, except that the severity of the headache had worsened. An epidural blood patch with 15 ml autologous blood injected through an 18-gauge Weiss needle inserted at L3–L4 was performed. The patient had minimal relief of her headache but refused further treatment. She was sent home with instructions to return if her headache worsened. The patient returned on day 3 complaining of severe headache with nausea and vomiting. A presumptive diagnosis of refractory PDPH was made. An intravenous infusion of Ringer’s lactate was begun secondary to nausea and vomiting and probable dehydration. Epidural blood patch was performed with 20 ml autologous blood injected at L4–L5 using an 18-gauge Weiss needle. The patient stated that her headache was significantly improved. She was sent home with analgesics and instructions to drink as much fluid as possible and to return if the headache worsened. On day 5 she returned, stating that her headache was worse. It remained occipital and postural in nature. She denied nuchal rigidity, nausea and vomiting, fever, or visual changes. Neurologic examination was normal. She was admitted to the hospital for further evaluation. Her evaluation included consultation with neurologists, hematologic testing for evidence of sepsis, and magnetic resonance imaging of the brain to rule out mass lesion or cortical vein thrombosis. All examinations (complete blood count, magnetic resonance imaging) were normal. The neurologists concurred with the diagnosis of PDPH. Intravenous caffeine, hydration, analgesics, amytriptyline, and muscle relaxants were administered with some transient relief. Over the next 48 h the patient’s headache increased in intensity despite all measures. A third epidural blood patch with 20 ml sterilely obtained autologous blood was performed at L4–L5 using an 18-gauge Weiss needle. The patient again stated that the headache was better and was discharged home. Eight days after the combined spinal–epidural analgesia was administered, the patient reported that the headache was worse and was now most intense when she was supine. Because of the patient’s complaint of increasing severity of the headache in the supine position and lack of response to epidural blood patch, cortical vein thrombosis was again considered. A repeat magnetic resonance imaging examination of the brain was obtained and was unremarkable.
Based on literature reports of success in treating refractory PDPH with ACTH, a decision was made to offer this steroid treatment to the patient. 1,2On day 10, the patient agreed to the treatment, and after informed consent was obtained, an intravenous infusion of cosyntropin 0.5 mg in 1 l Ringer’s lactate was administered over 8 h. The patient had complete relief of her symptoms. She was able to ambulate without recurrence of her headache. She was discharged the following day asymptomatic and remained asymptomatic at 3 days and 1 week later.
The typical PDPH is frontal–occipital and may radiate to the neck or shoulders. It is aggravated by sitting or standing and is lessened or relieved when the patient is in the supine position. Associated symptoms include photophobia, nausea and vomiting, tinnitus, deafness, and abducens nerve palsy. 3The differential diagnosis for PDPH includes meningitis, sinus headache, tension headache, cerebral hemorrhage, cerebral infarction, preeclampsia, migraine headache, and cortical vein thrombosis. 4,5Although preeclampsia and meningitis are included in the differential diagnosis of PDPH, these possibilities were unlikely because of the lack of hypertension, fever, or nuchal rigidity.
Our decision to use steroid treatment was based on correspondence by Collier, 1who described the use of 20 U long-acting ACTH administered intramuscularly to relieve PDPH. Because ACTH is not commonly available, Collier used a synthetic form of 1 mg ACTH administered intramuscularly in his treatment of PDPH. He speculated that ACTH may stimulate the adrenal gland to increase cerebrospinal fluid production and possibly also increase β-endorphin output. In a separate letter, Foster 2reported using 1.5 U/kg ACTH infused for >1 h in 1 or 2 l Ringer’s lactate solution for treatment of PDPH. Final relief was established 6–12 h after the infusion was completed. He reported that this treatment for PDPH was 70% effective, without reporting the number of cases used to determine this percentage.
Adrenocorticotropic hormone is not available in our pharmacy, but a similar drug, cosyntropin, is available. Cosyntropin shows the full corticosteroidogenic activity of natural ACTH. Cosyntropin is an α 1-24 corticotropin, a synthetic subunit of ACTH. It contains the first 24 of 39 amino acids of natural ACTH. It may be given as an intravenous infusion over a 4–8-h period to provide a greater stimulus to the adrenal glands. Doses of 0.25–0.75 mg have been used in clinical studies of adrenal function. 6Our patient was treated with an intravenous infusion of 0.5 mg cosyntropin administered in 1 l Ringer’s lactate over 8 h. She had complete relief of her headache without recurrence.
Adrenocorticotropic hormone stimulates the adrenal cortex to secrete glucocorticoids, mineralocorticoids, and weak androgens. ACTH activates adenylyl cyclase and increases intracellular cyclic adenosine monophosphate, which is the second messenger for most, if not all, of the effects of ACTH. Corticosteroids exert a number of indirect effects on the central nervous system. Most patients respond with mood elevation, which may impart a sense of well-being despite the persistence of underlying disease. In addition, corticosteroids profoundly alter the immune responses of lymphocytes, which is important in the antiinflammatory actions of the glucocorticoids. The risk of cosyntropin administration is low. Aside from the rare hypersensitivity reactions, the toxicity of ACTH is primarily limited to the increased secretion of corticosteroids. Cosyntropin is generally less antigenic than native ACTH and is thus the preferred agent for clinical use. 6
The positive result of our treatment of this patient with cosyntropin does raise many questions. Is it possible that the headache had simply run its course (11 days) and that it was coincidental that the cosyntropin gave positive results? If the basis for conservative treatment of PDPH is to increase cerebrospinal fluid production, and the possible benefit from cosyntropin is also to increase cerebrospinal fluid production, why would cosyntropin work and epidural blood patch fail? Could the positive results we experienced with cosyntropin be secondary to its antiinflammatory or mood-elevation effects? We have no clear answers to these questions. However, because the treatment of PDPH with cosyntropin is noninvasive and may be beneficial, we believe it should be considered in the treatment of PDPH. Future studies are warranted to determine precisely the role of synthetic ACTH analogs in the treatment of PDPH.