1339

important factors for the prognosis of yersinia infection in patients with beta-thalassemia, it is possible that intravenous antibiotic treatment including fluoroquinolones ± alone or in combination ± as soon as the first symptoms of yersiniosis occur may prevent yersinia sepsis and complications such as life-threatening ARDS, especially to splenectomized patients.

References 1. Gayraud M, Scavizzi M et al (1993) Antibiotic treatment of Yersinia enterocolitica septicemia: a retrospective review of 43 cases. Clin Infect Dis 17: 405±410 2. Hoogkamp-Korstanje JAA, de Koning J, Heesemann J (1988) Persistence of Yersinia enterocolitica in man. Infection 16: 81±85

)

P. Mavrommati × T. Hatzis ( ) Critical Care Department, Aghia Sophia Children's Hospital, Thevon & Levadias Street, Athens 115 27, Greece, Tel.: + 30 (1) 7 48 86 86, Fax: + 30 (1) 7 79 76 49

R. M. Calcroft G. Joynt

Metoclopramide improves gastric motility Accepted: 6 September 1999 Sir: Dr. Jooste and colleagues have shown that metoclopramide improves gastric emptying in a miscellaneous group of critically ill patients [1]. In common clinical practice, however, prokinetic drugs are considered a therapeutic option only in patients who demonstrate gastric intolerance, as demonstrated by phenomena such as large nasogastric aspirates, reflux, or vomiting. While we acknowledge that the relationship between large gastric aspirates and decreased gastric motility is still speculative [2], it would be much more relevant to assess the efficacy of metoclopramide in critically ill patients who have been demonstrated clinically to have impaired gastric emptying. The likelihood of these patients having markedly impaired gastric emptying of a magnitude sufficient to cause clinically detectable gastric intolerance is

uncertain, although we acknowledge they may have impaired gastric emptying relative to normals. Factors previously shown to be associated with impaired gastric emptying include being critically ill (Simplified Acute Physiology Score 9.5 ± 3), older age and female sex, opiate use [3], dopamine [4], and other drugs such as anticholinergics. Apart from opiate and inotrope use, the authors give little information regarding these factors. The paper by Dr. Jooste et al. therefore does not address the relevant clinical question: Does metoclopramide improve gastric emptying in patients with gastric stasis or gastric feed intolerance? Assuming that blood paracetamol levels accurately reflect gastric emptying, the magnitude of the difference between the two groups is not substantial when compared with the control group's standard deviation [metoclopramide group 1387.2 ± 560.4 (mean ± SD); saline group 994 ± 623.6]. While the authors have shown a statistically significant increase in the AUC 120 (area under the curve over 120 min) in patients given metoclopramide, we would question the clinical significance of this result. Lastly, in patients with documented gastric intolerance, we have noticed that the use of 1.5 g paracetamol is frequently associated (25 % of cases studied) with detectable serum levels of paracetamol 24 h after the first dose. We would be interested to know whether the authors experienced this problem, and, if so, how did they compensate for this phenomenon (which we believe is merely a consequence of prolonged gastric emptying). In conclusion, we believe that ªthe administration of intravenous metoclopramide improved gastric emptying in a heterogeneous group of critically ill patientsº may have been shown in this study, that ªmetoclopramide is a useful prokinetic drug in this patient populationª has yet to be demonstrated.

References 1. Jooste CA, Mustoe J, Collee G (1999) Metoclopramide improves gastric motility in critically ill patients. Intensive Care Med 25: 464±468 2. McClave SA, Snider HL, Lowen CC, McLaughlin AJ, Green LM, McCombes RJ et al (1992) Use of residual volume as a marker for entereal feeding intolerance: prospective blinded comparison with physical examination and radiographic findings. JPEN J Parenter Enteral Nutr 16: 99±105

3. Heyland DK, Tougas G, King D, Cook DJ (1996) Impaired gastric emptying in mechanically ventilated critically ill patients. Intensive Care Med 22: 1339±1344 4. Tarling MM, Toner CC, Withington PS, Baxter MK, Whelpton R, Goldhill DR (1997) A model of gastric emptying using paracetamol adsorption in intensive care patients. Intensive Care Med 23: 256±260

)

R. M. Calcroft ( ) × G. Joynt Prince of Wales Hospital Shatin, Hong Kong Tel.: + 852 (269) 62689 Fax: + 852 (269) 62689 email: [email protected]

C. A. Jooste J. Mustoe G. Collee

Reply Accepted: 6 September 1999 Sir: Thank you for the opportunity to reply to Drs. Calcroft and Joynt. The purpose of our study was to ascertain the prokinetic efficacy of metoclopramide in critically ill patients, who frequently suffer gastric impairment [1, 2]. Calcroft and Joynt suggest that impaired gastric emptying may be unimportant when it is clinically undetectable. However, clinical tests for impaired gastric emptying are insensitive [3]. The implicit suggestion that subclinically impaired emptying is irrelevant is speculative. We omitted the data on age and sex as, in a crossover study, these are not confounding variables. No patient received drugs known to influence gastric emptying other than those mentioned in the paper. It is contended in the letter that the difference in paracetamol absorption between the saline and metoclopramide groups is not substantial. This assertion does not take into account the sequence of randomisation. In those patients who received saline first, paracetamol absorption was less than the quoted mean of 994 mg ´ min/l in all cases. It was only in those who had already received metoclopramide that the absorption exceeded this mean following saline. As we speculated in the discussion, this may reflect a residual metoclopramide effect. We reiterate our conclusion that metoclopramide is a useful prokinetic in this pa-

1340

tient population. Metoclopramide improves gastric emptying in patients who have reduced gastric motility.

References 1. Dive A, Moulart M, Jonard P, Jamart J, Mahieu P (1994) Gastroduodenal motility in mechanically ventilated critically ill patients: a manometric study. Crit Care Med 22: 441±447 2. Tarling MM, Toner CC, Withington PS, Baxter MK, Whelpton R, Goldhill DR (1997) A model of gastric emptying using paracetamol absorption in intensive care patients. Intensive Care Med 23: 256±260 3. McClave SA, Snider HL, Lowen CC, McLaughlin AJ, Green LM, McCombes RJ, Rodgers L, Wright SA et al (1992) Use of residual volume as a marker for enteral feeding intolerance: prospective blinded comparison with physical examination and radiographic findings. JPEN J Parenter Enteral Nutr 16: 99±105

)

C. A. Jooste ( ) × J. Mustoe × G. Collee Department of Anaesthesia and Intensive Care, King's Colleqe Hospital, Denmark Hill, London SE5 9RS, UK Tel.: + 44 (171) 3463358 Fax: + 44 (171) 3463632

B. Riou M. L. Cittanova

An international review of HES Accepted: 30 August 1999 Sir: We read with interest the review of Treib et al. [1] regarding hydroxyethyl starches. We would like to make some comments about the renal effects of hydroxyethyl starches because we disagree with some of the information provided and the way this important topic was presented. Treib et al. stated that osmotic-like lesions were found in transplanted kidneys when starches had been used in the organ donor. However, it has been reported that these histological lesions are not comparable to those called osmotic lesions and which are observed after administration of many plasma expanders. Indeed, these starch-induced lesions involved both the

proximal and the distal renal tubules. Since starches have not been identified in the vacuoles, Treib et al. suggested that this indicates no direct relationship with hydroxyethyl starch. This argument is misleading, since other possible mechanisms may be involved in the occurrence of this toxic renal effect besides unchanged starch accumulation. The authors also concluded from Coronel et al. [2] that the hemodynamic status of the donor is still the main determinant of these lesions. The study of Coronel et al. [3] did not provide any argument to support it, and there is no reason to conclude that we must only consider this risk factor and neglect all other clinically relevant factors. Moreover, it is difficult to compare a prospective randomized study [4] and a retrospective study with historical controls [3]. Treib et al. concluded that it should be advisable to use starches as a plasma volume expander with caution in brain-dead donors. We strongly disagree with this point. First, we can use or not use starches in brain-dead patients, but we do not understand how to use them with caution. Second, there is a very low probability that a second randomized study will ever be performed. Indeed, very few randomized studies have been performed in brain-dead patients until now. Anyway, since ªprimum non nocereº should remain the rule in medicine, we consider that starches must be contraindicated in brain-dead patients, at least until evidence to the contrary is provided. This position has been advocated by a French expert panel [5]. The way the renal effects of starches has been presented [1] suggests that the problem is limited to kidney transplantation. We agree that initial evidence of the deleterious renal effect of starches has been provided in patients undergoing kidney transplantation [4, 6], but several case reports suggest that this problem could also occur in other patients [7, 8]. Moreover, we believe that these studies raise some important questions on the potential renal effect of starches. Are we sure that these lesions do not frequently occur in patients with renal diseases, renal ischemia, and in critically ill patients? We remain surprised at the weak reaction of regulatory agencies after the publication of these studies [4, 6]. Fresenius laboratories have recently cancelled a clinical study of the renal effects of starches in patients undergoing aortic surgery. Why? There is little evidence that not looking at a potential side effect of a drug may be really interesting for a pharmaceutical company. Even if regulatory agencies and pharmaceutical companies do not seem to be interested in responding to this important question, international researchers should try to obtain such an answer. In-

deed, new starches are or will be proposed in the future, and there is presently no proof that they are devoid of the same renal effects.

References 1. Treib J, Baron JF, Grauer MT, Strauss RG (1999) An international review of hydroxyethyl starches. Intensive Care Med 25: 258±268 2. Coronel B, Mercatello A, Martin X, Lefrançois N (1997) Hydroxyethylstarch and renal function in kidney transplant recipients. Lancet 349: 884 3. Coronel B, Laurent V, Mercatello A, Bret M, Colon S, Colpart JJ, Moskovtchenko JF (1994) L'hydroxyØthylamidon peut-il †ter utilisØ lors de la rØanimation des sujets en Øtat de mort cØrØbrale pour don d'organe? Ann Fr Anesth Reanim 13: 10±16 4. Cittanova ML, Leblanc I, Legendre C, Mouquet C, Riou B, Coriat P (1996) Effect of hydroxyethylstarch in brain-dead kidney donors on renal function in kidney-transplant recipients. Lancet 348: 1620±1622 5. Pottecher T (1998) RØanimation du sujet en Øtat de mort encØphalique en vue de prØl›vement d'organes. ConfØrence d'Experts organisØ par la SociØtØ Française d'AnesthØsie et de RØanimation en collaboration avec l'Etablissement Français des Greffes et la SociØtØ Française de Transplantation. Elsevier, Paris 6. Legendre C, Thervet E, Page B, Percheron A, Nol LH, Kreiss H (1993) Hydroxyethylstarch and osmotic-nephrosislike lesions in kidney transplantation. Lancet 342: 248±249 7. Waldhausen P, Kiesewetter H, Leipnitz G, Scielny J, Jung F, Baumbauer R, von Blohn G (1988) Durch Hydroxyethylstarch induzierte passagere Niereninsuffizienz bei vorbestehender glomerularer Schädigung. Acta Med Austriaca 18 [Suppl 1]: 52±55 8. Dickernamm MJ, Filipovic M, Schneider MC, Brunner FP (1998) Hydroxyethylstarch-associated transient acute renal failure after epidural anaesthesia for labour analgesia and caesarean section. Nephrol Dial Transplant 13: 2706

)

B. Riou ( ) × M. L. Cittanova Department of Anesthesiology and Critical Care, CHU PitiØ-Salp†tri›re, 47 Boulevard de l'Hôpital, 75651 Paris Cedex 13, France Tel.: + 33 (1) 42 16 22 51 Fax: + 33 (1) 42 16 22 69 email: [email protected]