SEARCH

SEARCH BY CITATION

Keywords:

  • tetanus;
  • intrathecal;
  • equine tetanus antitoxin;
  • human antitetanus immunoglobulin
  • tétanos;
  • intrathécal;
  • immunoglobulines antitétaniques humaines;
  • antitoxines antitétaniques équines
  • tetanus;
  • intratecal;
  • antitoxina equina de tétano;
  • inmunoglobulina humana antitetánica

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Background  Mortality caused by tetanus is still a serious health problem in developing countries. Apart from immunization, early treatment with equine antitetanus serum (ATS) or human tetanus immunoglobulin (TIG) is the real treatment that can avoid death. On pathophysiological grounds intrathecal administration would be preferred because of high concentrations of the antiserum in cerebrospinal fluid and thus around the nerve roots. Many studies concluded on its effectiveness whereas others did not find any superiority of this method. However, most of those studies were not random and/or had no sufficient weight.

Objective  To assess the efficacy of intrathecal therapy with ATS in neonates and adults.

Methods  Meta-analysis: Clinical trials were identified by searching Medline, the Cochrane library and Current Contents. Published randomized studies in English or French comparing intrathecal therapy and intramuscular therapy (IMS) were analysed with Revman, R, and Stata software. Treatment effects were evaluated by relative risk (RR) between intrathecal vs. intramuscular administration.

Results  A total of 942 patients were included in 12 trials, 484 in the intrathecal group and 458 in the intramuscular one. The combined RR of mortality for intrathecal vs. IMS was 0.71 (95% CI, 0.62–0.81). The superiority of intrathecal therapy also emerged when the analysis was performed in subcategories of both adults and neonates and for high and low dose of intrathecal serotherapy.

Conclusion  Intrathecal administration of ATS or TIG is more beneficial than intramuscular administration in the treatment of tetanus.

Introduction  La mortalité liée au tétanos continue d’être un sérieux problème de santé publique dans les pays en voie de développement. Hormis la vaccination, le traitement précoce par du sérum antitétanique équin ou par des immunoglobulines tétaniques humaines reste le traitement approprié pour éviter le décès. Du point de vue physiopathologique, l'administration intrathécale est préférable car celle-ci permet d'obtenir des concentrations élevées d'antitoxines dans le liquide céphalorachidien, donc à proximité des racines nerveuses. Plusieurs études ont démontré l'efficacité de ce type de traitement alors que d'autres n'y ont trouvé aucun avantage par rapport à l'administration intramusculaire. Cependant, la plupart des études n’étaient pas randomisées et/ou manquaient de puissance.

Objectifs  L'objectif de cette étude était d’évaluer l'efficacité de la thérapie intrathécale chez les nouveaux nés et les adultes par méta-analyse.

Méthodes  Les essais cliniques ont été identifiés en cherchant dans Medline, dans la librairie Cochrane, ainsi que dans Current Contents. Les études randomisées, publiées en Anglais ou en Français et comparant les thérapies intrathécales et intramusculaires ont été analysées en utilisant les logiciels Revman et Stata. Les effets des traitements ont étéévalués par analyse de la différence en risque relatif.

Résultats  942 patients ont été inclus dans 12 essais cliniques: 484 dans le groupe intrathécal et 458 dans le groupe intramusculaire. Le risque relatif global de mortalité entre la thérapie intrathécale et la thérapie intramusculaire était de 0,71, en faveur de la thérapie intrathécale (IC95%: 0,62–0,81). La supériorité de la thérapie intrathécale a également été démontrée lors de l'analyse de sous catégories d’âge et de doses.

Conclusion  L'administration intrathécale d'immunoglobulines antitétaniques humaines ou d'antitoxines antitétaniques équines est supérieure à l'administration intramusculaire dans le traitement du tétanos.

Antecedentes  La mortalidad por tétano continua siendo un problema de salud serio en países en vías de desarrollo. Aparte de la inmunización, el tratamiento temprano con suero equino antitetánico (SAT) o con inmunoglobulina humana antitetánica (IGT) es la intervención que puede evitar la muerte. En términos patofisiológicos, la administración intratecal es preferible, puesto que resulta en una alta concentración del antisuero en el líquido cefaloraquídeo (LCR), y por lo tanto alrededor de las raíces nerviosas. Muchos estudios concluyen sobre su efectividad, mientras que otros no han encontrado ninguna superioridad en este método. Sin embargo, la mayoría de estos estudios no estaban aleatorizados y/o no tenían suficiente peso.

Objetivo  Evaluar la eficacia de la terapia intratecal con suero antitetánico en neonatos y adultos.

Métodos  Metanálisis. Se identificaron ensayos clínicos mediante búsquedas en Medline, la Cochrane library y Current Contents. Se analizaron los estudios aleatorizados publicados en ingles o francés, que comparasen la terapia intratecal con la intramuscular, utilizando los paquetes informáticos de Revman, R y Stata. Los efectos del tratamiento se evaluaron por el riesgo relativo entre la administración intratecal vs. intramuscular.

Resultados  Se incluyeron un total de 942 pacientes provenientes de 12 ensayos, 484 en el grupo intratecal y 458 en el intramuscular. El riesgo relativo combinado de mortalidad para la terapia intratecal vs. intramuscular fue de 0.71 [95% CI, 0.62–0.81]. La superioridad de la terapia intratecal también fue patente cuando se realizó el análisis por subcategorías, tanto de adultos y de neonatos, como para dosis elevadas y bajas de seroterapia intratecal.

Conclusión  La administración intratecal de IGT o LCR es más beneficiosa que la administración intramuscular en el tratamiento del tétano.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Since Ildirim's study in 1967, which stated the efficacy of intrathecal immunotherapy (ITS), many other studies have been published evaluating its efficacy. He demonstrated that the plasmatic level of antitetanus serum (ATS) was <0.008 IU/ml after intramuscular administration of 6000 IU of human ATS whereas this level reached 1024 IU/ml when administered intrathecally (Ildirim 1967). In addition, when ATS is administered systemically, the large protein molecule cannot readily cross the blood brain barrier and neutralise the unfixed toxin already present in the central nerve system (Gupta et al. 1980). In 1979 Sedaghatian found that the mortality rate and duration of hospital stay were not significantly different in a group of intrathecal therapy vs. a group with intramuscular therapy (IMS) (Sedaghatian 1979). A study comparing i.v. and intrathecal serotherapy was also interrupted by Vakil in 1979 because no significant difference was found between the two groups (Vakil et al. 1979). On the other hand, Bandari showed that intrathecal serotherapy leads to higher mortality than IMS (Bhandari et al. 1980). Seven years later, Mongi concluded that ITS was superior to IMS in the treatment of neonatal tetanus (Mongi et al. 1987). Recently, Menon confirmed this statement (Menon & Mathews 2002).

In 1991 Abrutyn tried to settle these two opinions by conducting a meta-analysis and concluded that there was no superiority of intrathecal serotherapy over the other methods (Abrutyn & Berlin 1991). Unfortunately all clinical trials in this meta-analysis were not randomized. Moreover, there was an important heterogeneity between the studies so that caution must be taken.

Most of the time, the intramuscular way is preferred probably because of comfort and convenience for administration. Meanwhile, tetanus continues to kill in developing countries. For example one estimate suggests that the annual incidence is near 15 per 100 000 in Sub-Saharan Africa and 40–60 per 100 000 in India (Ribero & Gayon 2000). This information is of great public health importance because the worldwide burden of the disease is still large. To assess the efficacy of intrathecal therapy, we conducted a meta-analysis of randomized-controlled trials comparing intrathecal and intramuscular administration in the treatment of tetanus.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Trials were identified by searching the Medline database, the Cochrane library and the Current contents and other electronic data providers using following keywords: intrathecal tetanus antitoxin; intramuscular tetanus antitoxin; tetanus immunoglobulin (TIG). All articles published or translated in French or in English and matching our keywords were identified.

Next we selected only randomized trials focusing the comparison of intrathecal vs. intramuscular routes. Necessary data were: the country in which the trial was conducted; the year of publication; the sample size; in both two arms the age of the subjects, the dosage, the type of intervention and the mortality. Other criteria like the length of hospital stay, side effects of the treatment, length of follow-up were also identified but they did not contribute to exclude the trials.

The principal end point of our study was the evaluation of the efficacy of ITS vs. IMS by calculating the mortality rate within the two groups The secondary end point consisted in calculating the mortality rate in subcategories: adult vs. neonates; low doses vs. high doses; serum (horse) vs. human TIG.

We used Revman 4.2 software to record and to analyse data, further analyses were performed using the statistical software packages R 2.1 and Stata 8.2. Combining trials was performed using the Mantel–Haenszel method, the heterogeneity statistic being given by Cochrane Q as:

  • image

where the size of each trial θi is given the weight ωi in the analysis.

θMH is the overall estimate of the pooled effect, and it is given by: θMH=∑ωiθi/∑ωi, θ is the log relative risk (RR) or risk difference and the ωi are the weights calculated as 1/SE(θi)2 under the null hypothesis that there are no differences in treatment effect between trials. This follows a chi-squared distribution on k-1 degrees of freedom (where k is the number of studies contributing to the meta-analysis) and standard error (SE).

The statistic I2 was also calculated as I2 = (Q − (k − 1) × 100)/Q. This measures the extent of inconsistency among the studies’ results, and is interpreted as approximately the proportion of total variation in study estimates, i.e. because of heterogeneity rather than sampling error. The Z-test was also performed to assess the difference between the groups with significance level at 1.96. The null hypothesis of homogeneity was rejected at significance level of 0.05. Heterogeneity was also assessed using the Galbraith plot as suggested by Devillé, the standardized RR being given by lnRR/SE (y-axis) against the inverse of SE (x-axis) (Deville et al. 2002). The efficacy of the treatment was determined by calculating the RR. The impact of possible predictors on the treatment difference between intrathecal and intramuscular serotherapy was assessed using metaregression (van Houwelingen et al. 2002).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

In this meta-analysis, 12 clinical trials were selected involving 942 patients. Most of the trials were conducted in developing countries. Six trials involved adult subjects and six were conducted in neonates. The characteristics of the trials are shown in Tables 1 and 2. The dose in the ITS group was ranging between 50 and 1500 IU and between 250 and 40 000 IU in the IMS group. Figure 1 shows the Galbraith plot of the 12 studies. All the studies are within the 95% boundaries, meaning that there is homogeneity between studies. Figure 2 shows the RR for intrathecal serotherapy in individual trials and overall. The RR was below 1 in 11 out of the 12 trials and the overall RR was 0.71 (95% CI, 0.62–0.81, P-value < 0.001) meaning that the mortality rate was significantly lower for the intrathecal treatment.

Table 1.   Study characteristics of randomized trials comparing intrathecal with intramuscular administration of human immunoglobulin or equine antitoxin in the treatment of tetanus
AuthorsCountryYearAgeScoreSerumSteroidsMortality/sample sizeDoseMethods
TotalITSIMSITSIMSStudy type
  1. RT, randomized trial; SRSB, stratified randomized single blind; SRDB, stratified randomized double blind trial; ITS, intrathecal immunotherapy; IMS, intramuscular therapy; score, the Dakar score represents a scale of morbidity – 1 few symptoms, 2 moderate, 3 serious.

AgarwalIndia1998Adults1,3HumanNo14/367/187/18250750SRDB
ArunIndia1980Neonates2,3EquineYes54/9920/4934/5050–10010 000SRSB
AsimTurkey1980Neonates1,2,3EquineNo19/418/2111/20250500RT
ChughIndia1985Neonates1,2,3HumanNo81/10641/5540/51250250RT
GuptaIndia1980Adults1,2,3HumanNo32/971/4910/482501000RT
KeswaniIndia1980Adults1,2,3EquineNo49/11822/7027/4820010 000RT
MbiseTanzania1984Adults1,2,3EquineNo23/368/1815/18150010 000RT
Miranda FilhoBrazil2004Adults1,2,3HumanNo14/1204/5810/6210003000RT
MongiTanzania1987Neonates1,2,3EquineNo42/6615/3327/33150010 000RT
PaulIndia1984Neonates1,2,3EquineNo64/8732/4532/42250250RT
SandersEngland1977Adults1,2,3EquineYes35/7616/3819/382001500RT
SedhagatianTurkey1979Neonates1,2,3HumanNo28/6013/3015/3015040 000RT
Table 2.   Outcome, duration of hospital stay and follow-up in randomized trials comparing intrathecal with intramuscular administration of human immunoglobulin or equine antitoxin in the treatment of tetanus
AuthorsOutcomeHospital stayFollow-up duration
  1. ITS, intrathecal immunotherapy.

AgarwalAll mild tetanus in control group: evolution to severe tetanusSignificant difference in mild tetanus for hospital duration in favour of IT group (12.3 days) vs. IM group (24.1 days)Period of hospitalization
ArunNo difference in mortality rate according to the doseHospital stay 7.4 days in IT group; 10.4 IM groupPeriod of hospitalization
AsimMortality: 38% in treated group, 55% in control groupFollow-up: 6 months duration of therapy: 19.7 days in ITS and 20.4 days in control groupPeriod of hospitalization
ChughMortality: higher in control group than in IM and IT groupDuration in hospital: 3.34 days IT IM group, 3.31 days control group: 2.63 daysPeriod of hospitalization
GuptaMortality: higher IM group more progression to severe tetanus observed in IM groupDuration in hospital not reported2–3 months
KeswaniMortality: higher in control group than IT groupDuration in hospital not reportedPeriod of hospitalization
MbiseMortality higher in control group (IM) than IT groupDuration in hospital not reportedPeriod of hospitalization
Miranda MilhoBetter clinical outcome on 10th day in IT group, difference in recurrence of spasms or in hospital durationShorter duration of hospital stay in ITS groupPeriod of hospitalization
MongiMortality rate: intrathecal group = 45%, IM group = 82%19.3 days in IT group, 28.7 in IM groupPeriod of hospitalization
PaulMortality rate: 71% in IT group, 76% in IM groupDuration in hospital not reportedPeriod of hospitalization
SandersMortality: no significant difference observedDuration of hospital stay not reportedPeriod of hospitalization
SedhagatianMortality rate: no differenceDays in hospital: no difference, days in sedation: no differencePeriod of hospitalization
image

Figure 1.  Galbraith plot showing standardised treatment effect estimate [log relative risk divided by standard error (SE)] vs. precision (inverse of SE) of randomized trials comparing intrathecal with intramuscular administration of human immunoglobulin or equine antitoxin in the treatment of tetanus.

Download figure to PowerPoint

image

Figure 2.  Treatment effects expressed as relative risks of individual randomized trials and all trials combined comparing intrathecal with intramuscular administration of immunoglobulin or equine antitoxin in the treatment of tetanus (studies are ordered according to sample size).

Download figure to PowerPoint

Figure 3 presents a sensitivity analysis of the meta-analysis. Results from a random effects meta-analysis were consistent with those from the fixed effects model. In all subgroups, defined by the subject age, serum/Ig type, IT dose, geographical location, study size, and overall mortality rate, the intrathecal route showed a significantly lower mortality rate than the intramuscular route. There was a significantly (P = 0.037) higher reduction in mortality, compared with intramuscular administration, in studies using high dose ITS (>250 units) than in studies using low dose ITS (≤250 units). However, these results should be interpreted with care as only three studies used high dose ITS. Other factors did not show a significant effect on the difference in mortality between intrathecal and intramuscular route. The treatment difference appeared to be smaller in neonates, compared with adults. This difference was not statistically significant (P = 0.194), but could not be explained by confounding with ITS dosage, type of serotherapy, or overall mortality rate (Figure 4).

image

Figure 3.  Treatment effects of randomized trials comparing intrathecal with intramuscular administration of immunoglobulin or equine antitoxin in the treatment of tetanus, according to selected study characteristics (age, serum type, dose …).

Download figure to PowerPoint

image

Figure 4.  Treatment effects of randomized trials comparing intrathecal with intramuscular administration of immunoglobulin or equine antitoxin in the treatment of tetanus, combining age with dose, serum type and overall mortality rate.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

In this meta-analysis we found out that ITS was more effective than IMS. To minimize bias we selected only randomized trials as suggested by Cucherat et al. (1997). This contributed certainly to an increased strength of our study but some weaknesses should be pointed out: trials were not blinded (which is impossible with this type of treatment comparison), stratification before randomization according to the score of severity was not systematic in all trials, and dosage was different among trials. Moreover, some trials added corticotherapy to the serotherapy whereas others did not. These factors contributed probably to the reduction of homogeneity. The limitation of studies published in French or English was indicated by practical reasons, and probably did not affect the representativity of this meta-analysis. For studies in other languages with an abstract in English, only one met the inclusion criteria (Levin et al. 1987). All the studies were conducted in neonates and adults. No trial was performed in young children. This is probably because of the rarity of tetanus in this subcategory because of generalized worldwide programs of immunization at this age.

Several authors state that high doses of ITS could lead to better results than low doses (Mbise et al. 1984; Chopra et al. 1986; Mongi et al. 1987). Our analysis confirms this trend, although few studies used a high dose.

Results were not stratified by the severity of the disease because most of the trials included all stages of the tetanus without randomization for severity. However, without an effective treatment a mild tetanus can switch to severe tetanus and vice versa. Agarwal showed that intrathecal therapy reduces the progression of minor tetanus to severe tetanus whereas 50% of patients with minor tetanus who received intramuscular serotherapy developed severe tetanus within 24 h and 70% within 48 h (Agarwal et al. 1998).

We did not compare the duration of hospitalization between the groups because in several trials this item was lacking, and for the remainder it was often not possible to know if the duration in hospital was calculated in all patients or in survivors only.

Singh et al. showed that ITS was more effective than IMS when the serum is given within 24 h of onset of the convulsions (Singh et al. 1980). Many other authors confirmed this (Sanders et al. 1977; Singh et al. 1980; Mbise et al. 1984; Robert et al. 1984; Chopra et al. 1986; Mongi et al. 1987). Intrathecal immunotherapy could be difficult to perform but Gupta did not find any problem, patients being relaxed after i.v. diazepam (Gupta et al. 1980).

No neurological side effects of the intrathecal route were reported. However, in most of the trials the follow-up did not extend over the period of hospitalization. In fact, many authors claim that the intrathecal route for serotherapy is innocuous (Diop-Mar et al. 1975). Gupta did not find any definite neurological deficit after ITS during 3 months of follow-up (Gupta et al. 1980). On the other hand Robert et al. reported reversible paraplegia after high dose of intrathecal human immunoglobulin (Robert et al. 1984). This complication had been attributed to either the high dose of immunoglobulin or to drug preservatives.

Do other aspects of management of tetanus play an important role in the prognosis? Recent advances in intensive care especially those concerning spasms and neurovegetative manifestations and respiratory distress management should probably explain the reduction of mortality observed in recent trials like in the Miranda study but not in old ones like the Gupta study (Sutton et al. 1990; Attygalle & Rodrigo 2002, 2004; Reddy 2002; Thwaites et al. 2004). However, other studies showed that even in well-equipped intensive care units the mortality because of the tetanus remains very high (Attygalle & Rodrigo 2004). Hyperbaric oxygen is commonly and successfully used in the treatment of Clostridium perfringens infections. We did not find any trial in English or in French stating that this treatment was effective in tetanus cases. It might be that authors did not find any benefit in this method, or that a randomized trial was judged unethical.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

We conclude that intrathecal serotherapy is more effective than intramuscular serotherapy and should be preferred whenever tetanus is suspected. The difference is slightly higher in adult subjects and significantly higher when using high dose ITS (dosage >250 units).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  • Abrutyn E & Berlin JA (1991) Intrathecal therapy in tetanus. A meta-analysis. JAMA 266, 22622267.
  • Agarwal M, Thomas K, Peter JV, Jeyaseelan L & Cherian AM (1998) A randomized double-blind sham-controlled study of intrathecal human anti-tetanus immunoglobulin in the management of tetanus. The National Medical Journal of India 11, 209212.
  • Attygalle D & Rodrigo N (2002) Magnesium as first line therapy in the management of tetanus: a prospective study of 40 patients. Anaesthesia 57, 811817.
  • Attygalle D & Rodrigo N (2004) New trends in the management of tetanus. Expert Review of Anti-Infective Therapy 2, 7384.
  • Bhandari B, Ajmera NK & Jagetiya PR (1980) Intrathecal anti-tetanus serum in management of tetanus neonatorum. The Indian Journal of Medical Research 72, 685687.
  • Chopra K, Gupta A & Mhatre K (1986) Intrathecal tetanus hyperimmune human gamma globulin in the treatment of tetanus. Indian Pediatrics 23, 775778.
  • Cucherat M, Boissel JP & Leizorovicz A (1997) La méta analyse des essaies thérapeutiques, Masson, Paris.
  • Deville WL, Buntinx F, Bouter LM et al. (2002) Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Medical Research Methodology 2, 9.
  • Diop-Mar I, Badianes S, Diague S, Stellmann C & Rey M (1975) Confirmation de l'efficacité de la sérothérapie intrathécale par une étude randomisée de 245 cas. Compte rendu de la 4ième conférence internationale sur le tétanos, Dakar, Senegal, pp. 641645.
  • Gupta PS, Kapoor R, Goyal S, Batra VK & Jain BK (1980) Intrathecal human tetanus immunoglobulin in early tetanus. Lancet 2, 439440.
  • Van Houwelingen HC, Arends LR & Stijnen T (2002) Advanced methods in meta-analysis: multivariate approach and meta-regression. Acta Genetica et Statistica Medica 21, 589624.
  • Ildirim I (1967) A new treatment for neonatal tetanus. Antitetanic serum and prednisolone given together intrathecally. The Turkish Journal of Pediatrics 9, 8995.
  • Levin AS, Barone AA & Shiroma M (1987) Intrathecal serotherapy in tetanus (review). Revista do Instituto de Medicina Tropical de Sao Paulo 29, 255267.
  • Mbise RL, Mongi PS, Msengi AE & Amsi DO (1984) Tetanus neonatorum – experience with intrathecal serotherapy at Muhimbili Medical Centre Dar es Salaam. East African Medical Journal 61, 892895.
  • Menon J & Mathews L (2002) Intrathecal immunoglobulin in the treatment of tetanus. Indian Pediatrics 39, 654657.
  • Mongi PS, Mbise RL, Msengi AE & Do Amsi DM (1987) Tetanus neonatorum – experience with intrathecal serotherapy at Muhimbili Medical Centre, Dar es Salaam, Tanzania. Annals of Tropical Paediatrics 7, 2731.
  • Reddy VG (2002) Pharmacotherapy of tetanus – a review. Middle East Journal of Anesthesiology 16, 419442.
  • Ribero R & Gayon A (2000) Le traitement du tétanos en zones rurales de l'Afrique de l'Est (RDC). Evaluation thérapeutique à propos de 21 cas. Médecine d'Afrique Noire 47, 141147.
  • Robert R, Rouffineau J, Cremault A et al. (1984) Paraplégie réversible aprè injection intrathécale de fortes doses de gammaglobulines humaines lors du traitement de tétanos de faible gravité. Quatre observations. La Presse Médicale 13, 19471979.
  • Sanders RK, Martyn B, Joseph R & Peacock ML (1977) Intrathecal antitetanus serum (horse) in the treatment of tetanus. Lancet 1, 974977.
  • Sedaghatian MR (1979) Intrathecal serotherapy in neonatal tetanus: a controlled trial. Archives of Disease in Childhood 54, 623625.
  • Singh AK, Bansal A, Goel SP & Agarwal VK (1980) Intrathecal antitetanus serum (horse) with steroid in the treatment of neonatal tetanus. Archives of Disease in Childhood 55, 527531.
  • Sutton DN, Tremlett MR, Woodcock TE & Nielsen MS (1990) Management of autonomic dysfunction in severe tetanus: the use of magnesium sulphate and clonidine. Journal of Intensive Care Medicine 16, 7580.
  • Thwaites CL, Yen LM, Nga NT et al. (2004) Impact of improved vaccination programme and intensive care facilities on incidence and outcome of tetanus in southern Vietnam, 1993–2002. Transactions of the Royal Society of Tropical Medicine and Hygiene 98, 671677.
  • Vakil BJ, Armitage P, Clifford RE & Laurence DR (1979) Therapeutic trial of intracisternal human tetanus immunoglobulin in clinical tetanus. Transactions of the Royal Society of Tropical Medicine and Hygiene 73, 579583.