How Can We Help?
Other CNS infections
Suspected encephalitis.
Consider this in patients presenting with headache, fever, and altered consciousness, and the finding of focal neurological signs or the presence of focal fits should increase the probability of the illness still further. In high HIV seroprevalence areas chronic lymphocytic meningitides due to tuberculosis, cryptococcosis or even syphilis need to be considered, because all three can cause a vasculitis with associated stroke-like features.
Serious consideration should be given to the following causes: herpes; rabies; HIV seroconversion illness, mumps and polio. polio1 Herpes is one of the most important to consider, as it is potentially treatable – consider it especially if there are temporal lobe features such as aphasia or temporal lobe fits.
Herpes encephalitis.
A CT scan is usually indicated, and may show temporal lobe changes. If it is considered safe to do an LP, the CSF usually shows some lymphocytes with a raised protein, although it may rarely be normal. The diagnosis can be made by PCR of the CSF. Treatment is with acyclovir at a dose of 10 mg/kg every 8 hours, given as an infusion over one hour (e.g. 500 – 750 mg 3x/d). Mortality has purportedly dropped from 70% to 20%2after it started being used, although controlled trials are difficult to find, and many of the surviving patients are left with some neurological sequelae.
Brain abscess.
This should be considered a surgical or neurosurgical condition, with the ideal being a burr hole and pus aspiration for microscopy and culture. Lack of fever and lack of a raised CRP3 do NOT rule out the diagnosis, which should be considered in any patient presenting with new onset headache, changing mental status or new seizures.
Antibiotic choice should be determined by culture results, but in their absence use ceftriaxone 2g 12 hourly and metronidazole 500 mg 8 hourly. Add cloxacillin 2g 6 hrly if there was head trauma. Evidence that brain abscesses can be managed safely without surgery is scanty, and limited mainly to retrospective cohorts; in patients with background meningitis, multiple abscesses, abscesses in neurologically sensitive areas, or abscesses where the likely organism has been identified in another way, medical management may be reasonable.4
Neurocysticercosis
Neurocysticercosis is readily identified on CT, but the management is still controversial.
CT findings in neurocysticercosis
One or more cystic (‘water-filled’) lesions containing a small hyper-dense mass is practically pathognomonic. As the cyst dies, an immune response occurs, with an area of oedema seen round the cyst. Finally the remains of the cyst calcifies.
Racemose (extraparenchymal) neurocysticercosis
This can be tricky, and easily missed on CT although usually identified on MRI. Cysts can be meningeal, intraventricular, or subarachnoid, and can present with meningeal features, hydrocephalus, mass effects, cerebrovascular disease, and even dementia.5 Management is linked to presentation, but initial corticosteroid therapy is probably important before using albendazole, particularly if there is cerebrovascular involvement.
Treatment of neurocysticercosis
Calcified cysts require no therapy, even if numerous. Treat the epilepsy.
Dying cysts with surrounding oedema should not be treated with a cysticidal medication initially, as the heightened inflammatory response may lead to clinical deterioration with status epilepticus or reduced consciousness due to encephalopathy.6 Prednisone 1 mg/kg (or equivalent dose of dexamethasone) is appropriate if the patient is symptomatic.
Viable cysts without oedema are arguably the lesions most amenable to therapy, and recommendations for albendazole 15 mg/kg (e.g. 400 mg 12 hourly for an individual weighing more than 60 kg) per day for 3 to 28 days abound. It is enormously tempting to treat such patients, and most clinicians succumb, although the evidence that doing so really benefits the patient is quite scanty. If giving albendazole, 10 days is reasonable, and combine it with dexamethasone 4 mg 12 hourly for the duration (or prednisone 60 mg/day).
If you have given treatment, it is traditional to do another scan after a few months to ‘assess response’.
Perspective – antiparasitic treatment in neurocysticercosis.
A Cochrane review7 found no evidence of benefit in terms of fit reduction, headache relief, disability or death, but phrased the conclusion as ‘insufficient evidence to demonstrate benefit.’ Subsequently, another moderate sized (60 patients in each arm) trial8 demonstrated both a reduction in parasite burden and in fits with generalisation. On the face of it, this seems quite convincing, but it is worth looking into more closely:
The mechanism of patient recruitment is unclear, and unfortunately could be a source of bias. The main difficulty is the use of two active agents – the patients in the treatment arm were given both albendazole (400 mg 12 hourly for 10 days) and dexamethasone (2 mg 8 hourly also for 10 days), whereas those in the placebo arm received neither. The primary endpoint was changed from ‘all seizures’ to ‘seizures with generalisation.’ It is unclear from the report whether the neurologists following up the patients were blinded. (The patients were followed up by neurologists and not by generalists or medical officers.) There was considerable discrepancy between the actual and expected seizure rates in the placebo group – the cumulative incidence at 30 months was predicted to be 70%. In practice, 49% of patients in the placebo group had at least one seizure compared with 44% in the active group:
Endpoint | Placebo | Active | ARR | NNT |
N | 59 | 57 | ||
No fits | 50.8% | 56.1% | 5.3% | 19 (NS) |
Partial fits | 27.1% | 33.3% | -6.2% | 17(NNH!) |
Generalised fits | 37.3% | 22.8% | 14.5% |
|
(The trial can be read in a more kindly manner by looking at the total number of partial fits, rather than the number of patients experiencing them, in which case albendazole is associated with a non-significant reduction in partial seizures.) The conclusion from all this is that it is still unclear how real the actual clinical benefit to patients of giving albendazole is, although there are tempting glimpses from underpowered trials in selected patients to suggest modest efficacy. The end result is a 2021 review still unable to determine clear benefit9
Tetanus.
Septic circumcision is still a common cause, but any contaminated wound, even a relatively minor puncture, may be sufficient. Be alert for minor localised forms that are variously labelled as epilepsy, dystonias, or even, in the case of abdominal presentation, appendicitis. The generalised, inducible, arching, rigid spasms of severe tetanus are highly characteristic and once seen should be instantly recognisable in future patients.
Management
This should ideally be in an ICU unless very mild.
Airway. Early tracheostomy is important. Trying to intubate somebody with ongoing spasms without adequate sedation and muscle relaxants at hand is very difficult – ask for assistance from an anaesthetist if in any doubt. Also, a properly performed tracheostomy done electively a day or so later is probably preferable to one performed under duress by a non-expert.
Immunisation and immunoglobulin: Tetanus hyperimmune globulin 3000 units IMI (intrathecal is of no extra benefit and the spasms make lumbar puncture technically difficult). The preparation comes as either 250 IUml or 500 IU/ml; it is recommended to divide the dose if the volume given is more than 5 ml and inject into separate sites with no more than 5lm per site. There is some evidence that immunity to tetanus, even after a symptomatic episode, is incomplete, so don’t neglect to give tetanus toxoid 0.5 ml IM as well (on admission, at one month, and again at six months).
Immediate wound debridement. This is often a major issue, particularly when the tetanus is related to a septic circumcision, as there may be considerable reluctance on the part of the surgeons to perform ‘unnecessary’ surgery. The outcome without adequate debridement is uniformly poor.
Antibiotic: Penicillin G 5 MU 6 hourly or metronidazole 500 mg 8 hourly) IVI for 10 days.
Avoid unnecessary stimuli (noise, unnecessary venipunctures)
Sedation: diazepam at 5-10 mg per hour once ventilated. One may also need chlorpromazine 50 mg 4-hourly. Use NGT route for some of the sedation. Doses as high as diazepam 100 mg 2 hourly and chlorpromazine 200 mg 4 hourly have been used. Full paralysis with muscle relaxants is often unnecessary, but agents such as pancuronium or vecuronium may be helpful.
Analgesia with paracetamol 1g 6 hourly may be adequate, but also consider morphine.
Autonomic instability is sometimes a problem – watch out for it and don’t over-react to each fluctuation in pulse – you may end up with a patient on a cocktail of drugs with no way of knowing whether the subsequent profound hypotension is due to the disease or the doctor. There is anecdotal and small study evidence about a vast array of magic medicines that cure this condition. Be careful. In adequately sedated patients, this condition is rare, but in poorly sedated patients where spasms are being dealt with primarily with muscle relaxants it may be problematic.
Adequate enteral nutrition is essential, and should be commenced as soon as the airway is secured and aspiration can be prevented safely.
Also ensure adequate fluid intake, and giver heparin for DVT prophylaxis.
The natural history of the condition is toward recovery once the infection is removed, but it will take weeks rather than days, so be patient.
Perspective – intrathecal immunoglobulin
Small studies on the value of intrathecal immunoglobulin in treating tetanus continue to appear with predictable regularity. Most are underpowered to reach any conclusions, and yet on superficial perusal they often appear to show dramatic benefit. An example was published in the BMJ.10 This trial of 120 patients showed dramatic ARRs for spasm duration and length of hospital stay, although no survival advantage. Closer scrutiny reveals that the caregivers were not formally blinded, the treatment environment changed during the study (they opened an ICU) and the placebo group got 25% less immunoglobulin than the active treatment group. It is unlikely that the addition of this study will change the conclusion from earlier reviews that there was no benefit from intrathecal therapy.
Perspective – penicillin or metronidazole
A tiny study (unblinded and not randomised) from Indonesia showing a mortality advantage of metronidazole over penicillin has been long touted as evidence for using the former, as it is proposed that Pen G is a GABA inhibitor which may potentiate the effect of tetanus toxin.11
A subsequent similarly small three armed study compared benzathine penicillin (1.2 MU as a single dose), Pen G, and oral metronidazole12 in a randomised controlled trial that was also rather small (161 patients) and found no differences in outcomes. It is moot whether any antibiotic treatment is actually likely to change outcomes so choice is likely immaterial other than potentially generating resistance with the longer courses of Pen G and metronidazole.
Whitley RJ, Gnann JW. Viral encephalitis: familiar infections and emerging pathogens. Lancet. 2002;359:507-14 ↩
McGrath N, Anderson NE, Croxson MC, et al. Herpes simplex encephalitis treated with acyclovir: diagnosis and long term outcome. J Neurol Neurosurg Psychiatry. 1997;63:321-6 ↩
Oyama H, Kito A, Maki H, Hattori K, Noda T, Wada K. Inflammatory index and treatment of brain abscess. Nagoya J Med Sci. 2012 Aug;74(3-4):313-24. PMID: 23092104; PMCID: PMC4831240 ↩
S.-Y. Hsiao, W.-N. Chang, W.-C. Lin, N.-W. Tsai, C.-R. Huang, H.-C. Wang, C.-M. Su, M.-J. Chuang, C.-H. Lu,The experiences of non-operative treatment in patients with bacterial brain abscess,Clinical Microbiology and Infection,Volume 17, Issue 4,2011,Pages 615-620,ISSN 1198-743X,https://doi.org/10.1111/j.1469-0691.2010.03264.x. https://www.sciencedirect.com/science/article/pii/S1198743X14632845 ↩
Mahale RR, Mehta A, Rangasetty S. Extraparenchymal (Racemose) Neurocysticercosis and Its Multitude Manifestations: A Comprehensive Review. J Clin Neurol. 2015 Jul;11(3):203-11. doi: 10.3988/jcn.2015.11.3.203. Epub 2015 May 28. PMID: 26022457; PMCID: PMC4507373. ↩
Garg RK, Uniyal R, Malhotra HS. Be careful while using albendazole/praziquantel in neurocysticercosis.Neurol India 2017;65:924-926 ↩
Salinas R, Prasad K. Drugs for treating neurocysticercosis. In: The Cochrane Library, Issue 2, 2002. Oxford: Update Software ↩
Garcia HH, Pretell EJ, Gilman RH, et al. A trial of antiparasitic therapy to reduce the rate of seizures due to cerebral cysticercosis. N Engl J Med 2004;350:249-58. [1] Vazquez VA, Sotelo J. The course of seizures after treatment for cerebral cysticercosis. N Engl J Med. 1992;327:696-701 ↩
Monk EJM, Abba K, Ranganathan LN. Anthelmintics for people with neurocysticercosis. Cochrane Database of Systematic Reviews 2021, Issue 6. Art. No.: CD000215.DOI: 10.1002/14651858.CD000215.pub5 ↩
Miranda-Filho DB, Ximines RA, Barone AA, et al. Randomised controlled trial of tetanus treatment with antitetanus immunoglobulin by the intrathecal or intramuscular route. BMJ 2004;328:615-7 ↩
Ahmadsyah I, Salim A. Treatment of tetanus: an open study to compare the efficacy of procaine penicillin and metronidazole. British Medical Journal (Clinical Research ed.). 1985 Sep;291(6496):648-650. DOI: 10.1136/bmj.291.6496.648. PMID: 3928066; PMCID: PMC1417474. ↩
Ganesh Kumar AV, Kothari VM, Krishnan A, Karnad DR. Benzathine penicillin, metronidazole and benzyl penicillin in the treatment of tetanus: a randomized, controlled trial. Ann Trop Med Parasitol. 2004;98:59–63. doi: 10.1179/000349804225003037 ↩