35 cmH2O with a PEEP of 5 cmH2Oare considered is 210 ml, while 300 mlis set on the ventilator [2].

2. The second issue, which may possiblyexplain the results reported by Zanettaet al. is related to the technical limita-tion of the Venturi inspiratory flow sys-tems. With these systems, inspiratoryflow may be generated either at 100%or at 60% FIO2. Nevertheless, at 60%FIO2, the pressurisation gas source maybe insufficient to overcome a high re-sistive load, subsequently resulting inan increased FIO2 and a reduction inthe flow delivered [3]. This phenome-non is particularly marked when the setinspiratory flow is low. Moreover, thisis interestingly resolved by switchingFIO2 from 60 to 100% FIO2. Figure 1illustrates the switch from a FIO2 of100% to 60%, with a Vt set at 300 mland a given resistance of 50 cmH2O.s/l.In this experiment, the fall in peak pres-sure clearly suggests the weakness ofthe Venturi flow delivery system.

Implication of these findingsin clinical practice

Although the ability of transport ventilatorsto deliver set Vt may be a relevant issue inpatients exhibiting high resistive load, infact PEEP and FIO2 are at least as impor-tant to guarantee oxygenation in the major-ity of critically ill patients. For the purposeof limiting volume loss, increasing the setVt according to the expected compressedVt, as well as the use of 100% FIO2 may

be safely recommended in patients com-bining low inspiratory Vt and high resis-tive load.

Whether a sophisticated ICU ventilator,that was not specifically designed fortransport, should be used compared to asimpler transport ventilator remains con-troversial and warrants further investiga-tion.

References

1. Zanetta G, Robert D, Guérin C (2002)Evaluation of ventilators used duringtransport of ICU patients – a benchstudy. Intensive Care Med 28:443–451

2. McGough EK, Banner MJ, Melker RJ(1992) Variations in tidal volume withportable transport ventilators. RespirCare 37:233–239

3. Kacmarek RM, Hess DR (1993) Airwaypressure, flow and volume waveforms,and lungs mechanics during mechanicalventilation. In: RM Kacmarek, DRHess, JK Stoller (eds) Monitoring in res-piratory care. Mosby, Saint-Louis,pp 497–543

L. Breton · G. Minaret · J. AboabJ.-C. Richard (✉ )Medical Intensive Care Unit, Rouen University Hospital, 1, Rue de Germont, 76000 Rouen, Francee-mail: jrichard@invivo.eduTel.: +33-232-888261Fax: +33-232-888314

Intensive Care Med (2002) 28:1181DOI 10.1007/s00134-002-1390-7 C O R R E S P O N D E N C E

Lucie BretonGuillaume MinaretJérôme AboabJean-Christophe Richard

Fractional inspired oxygenon transport ventilators: an important determinantof volume delivery during assistcontrol ventilation with high resistive load

Received: 15 May 2002Accepted: 30 May 2002Published online: 5 July 2002© Springer-Verlag 2002

Sir: Zanetta et al. reported an interestingbench test study dealing with the specificproblem of ventilators designed for trans-port [1]. The authors showed that the per-formance of transport and intensive care(ICU) ventilators, set in a volume-con-trolled mode and submitted to variouscombinations of resistive and elastic loads,were very inhomogeneous. These findingsare of high clinical relevance, since theloss of tidal volume (Vt) they observedreached 30% of the set Vt when high resis-tance was considered.

As discussed by the authors, two mainmechanisms may explain, at least in part,these findings: (1) gas compression in theventilatory circuit; (2) technical limits ofthe Venturi flow delivery systems. A betterunderstanding of the respective role playedby these two features may permit the userto limit significantly the clinical impact ofproblems related to Vt delivery.

1. Except for the Horus (Taema, Antony,France), no ventilator investigated inthis study compensated for gas com-pression. This might explain the excel-lent performance exhibited by the Horus,whatever the condition tested, com-pared with both portable and other ICUventilators. Nevertheless, increasing theset Vt according to a theoretical amountof gas compressed in the circuit may ef-ficiently prevent this problem. In fact,when the compliance of the circuit isknown, the following equation allowsfor the calculation of the volume loss[Ccircuit*(Ppeak–PEEP), where Ccircuit iscircuit compliance, Ppeak peak pressureand PEEP positive end-expiratory pres-sure]. For instance, the volume effec-tively inflated into the lungs when aCcircuit of 3 ml/cmH2O and a Ppeak at

Fig. 1 Inspiratory flow and airway pressure traces for a tidal volume (Vt) set at 300 ml ona ventilator (Osiris 1, Taema, Antony, France) and an imposed resistance (R) at50 cmH2O.s/l during the FIO2 switch from position 100% (O2) to 60% (Air-O2). Thechange in FIO2 resulted in the peak pressure fall, due to the reduction in flow delivery. Inthis example, the volume reduction reached approximately 30%