The SARS-CoV-2 acute lung infection can induce severe hypoxia, which seems related to intense pulmonary blood vessel dilatation and severe intrapulmonary oxygen shunt without major respiratory mechanics alteration [1]. Some non-ventilatory strategies have been proposed:
-
•
ECMO is limited to trained centres, and cannot meet the vast COVID-19 pandemic demand;
-
•
the prone positioning, supposed to facilitate alveolar recruitment and decrease the heterogeneity of compliance, appeared a good option [2] and improved PaO2 was frequently observed;
-
•
in the 90s’, pharmacological strategies have been reported, especially almitrine bismesylate [3].
According to the French National agency for Drug Security (ANSM), only IV almitrine is indicated for hypoxic acute respiratory failure as a Drug of Major Therapeutic Interest. The hypothesis was that almitrine might restore oxygenation, even partially, both in supine or prone positioning in a case series of mechanically ventilated COVID-19 patients during their early phase.
1. Patients and measurements
The Research Program was approved by the Direction de la Recherche et de l'Innovation (DRI) (ref. 2020PI080), by the research Ethical Committee (Saisine 263) of the Centre Hospitalier Régional Universitaire (CHRU) de Nancy, France, and registered at ClinicalTrials.gov (NCT04380727). The relatives or patients were questioned about objections to use collected data for scientific purposes and/or potential publications. The statement and objection form were dated and recorded in the medical file.
Between March 16th and April 12th, 2020, COVID-19 patients referred to ICU with positive PCR testing received IV almitrine bismesylate (Vectarion®, Servier Laboratory, Neuilly, France) for acute hypoxia in acute respiratory failure. After baseline measurement in prone or supine positioning, a second measurement was carried out 45 min after 4 mcg/kg/min, followed by 12 mcg/kg/min almitrine infusion rate, to test a dose -effect response. Matched (on age, BMI, gender, and baseline PaO2/FiO2) controls COVID-19 patients were compared over similar delays (8 hours H8). Patients having an acute cor pulmonale (trans-thoracic 2D Echo–Doppler) or abnormal liver function tests or hyperlactatemia [4] were not included. Data were reported as median (interquartile) for continuous variables and as count (percentage) for categorical variables. Changes during almitrine infusion were assessed using a Friedman test followed by a post hoc Wilcoxon signed-ranks test if appropriate. A two-point comparison was performed using a Wilcoxon signed-ranks test: baseline versus the best PaO2/FiO2 obtained with 4 or 12 mcg/kg/min almitrine infusion (almitrine group), and baseline versus H8 (matched control group). Tests were 2-sided and a P-value < 0.05 was considered as statistically significant. Statistical analyses were performed using R version 3.6.0 for Mac OS (The R Foundation for Statistical Computing, Vienna, Austria).
2. Results
Seventeen patients were included (10 males, age = 70 [54–78] years, BMI = 29 [23–34]) having chronic metabolic and cardiovascular diseases. Haemoglobin was stable at 12.2 [8.9–13.8] g/dL. All patients were intubated just before or soon after ICU admission and ventilated at FiO2 1. The length of stay in ICU was 18 [7–33] days. Four patients (24%) died in ICU.
The Fig. 1 showed individual data for ScvO2 (not shown in control group) and PaO2/FiO2 in both groups, which significantly increased only in the almitrine group. Eight over 10 patients increased their PaO2 with almitrine, with no clear relation with infusion rate. During this short perfusion time, no haemodynamic side effects were observed (Table 1 ) and lactate remained < 1.5 mmol/L. Only one of these severe patients underwent rescue ECMO (persistent PaO2/FiO2 < 70 mmHg).
Fig. 1.
Individual change of the PaO2/FiO2 ratio (A) and the ScvO2 (B) between baseline and post-almitrine administration (best dose response in terms of PaO2/FiO2 ratio) in the almitrine group of patients (n = 10). Individual change of the PaO2/FiO2 ratio (C) in the matched-control group (n = 7) between baseline and H8. ScvO2 measurements were not available for the control group. Shape of the individual point corresponds to the position of the patients (circle: prone positioning; triangle: supine positioning). Tukey boxplots on the background show correspondent median with 25th and 75thpercentiles (lower and upper hinges). Whiskers extend from the correspondent hinge to the largest or smaller value not further than 1.5*interquartile range.
Table 1.
Haemodynamic, pulmonary gas exchange and mechanical ventilation parameters at baseline and after 4 and 12 μg/kg/min of almitrine.
| Median (IQR) | Baseline | 4 μg/kg/min | 12 μg/kg/min | P |
|---|---|---|---|---|
| n | 10 | 10 | 10 | |
| FiO2 | 1.00 | 1.00 | 1.00 | NA |
| Positioning | ||||
| Prone | 3 | 3 | 3 | NA |
| Supine | 7 | 7 | 7 | |
| PaO2(mmHg) | 135 (85, 195) | 149 (91, 28) | 215 (121, 275) | 0.06 |
| PaCO2 (mmHg) | 42 (40, 45) | 42 (40, 45) | 43 (37, 46) | 0.45 |
| pH | 7.38 (7.36, 7.44) | 7.40 (7.36, 7.42) | 7.41 (7.36, 7.42) | 0.70 |
| HCO3 (mmol/L) | 25.1 (23.5, 28.2) | 25.3 (24.1, 27.9) | 24.8 (23.9, 27.6) | 0.91 |
| BE | 1.4 (−1.4, 2.7) | 1.1 (−0.3, 2.4) | −0.3 (−1.5, 2.5) | 0.91 |
| ScvO2 (%) | 73 (72, 76) | 81 (79, 82) | 85 (78, 87) * | 0.03 |
| Pra (mmHg) | 8 (7, 9) | 9 (8, 11) | 9 (8, 11) | 0.33 |
| CI (L/min/m2) | 2.1 (1.9, 2.2) | 2.2 (1.9, 2.3) | 2.2 (1.9, 2.5) | 0.42 |
| Vt (mL) | 425 (398, 465) | 425 (398, 465) | 425 (398, 465) | 1 |
| RR (/min) | 23 (22, 24) | 24 (22, 25) | 24 (22, 25) | 1 |
| PEEP (cmH2O) | 10 (10, 11) | 10 (8, 10) | 10 (10, 10) | 0.14 |
| PIPressure (cmH2O) | 27 (26, 29) | 27 (26, 30) | 27 (26, 29) | 1 |
| Pplat (cmH2O) | 22 (19, 24) | 22 (20, 24) | 22 (20, 24) | 1 |
BE: base excess; ScvO2: central venous O2 saturation; Pra: right atrial pressure; CI: cardiac index; CO: cardiac output; Vt: tidal volume; RR: respiratory rate; PEEP: positive end-expiratory pressure; PIPressure: peak inspiratory pressure; Pplat: plateau pressure; *: P < 0.05 vs. baseline value.
3. Discussion
Intravenous almitrine was associated with almost a doubling of the PaO2/FiO2 ratio in the early phase of severe COVID-19 acute respiratory failure with no dose-effect relationship. ScvO2 increased consistently, while neither Pra nor CI was altered. The individual increase in PaO2 to almitrine infusion varied in amplitude between patients, on average 80 mmHg. Just like others, we have beforehand shown that almitrine may spectacularly improve PaO2 by reducing intrapulmonary oxygen shunt [3], [5]. Among other non-ventilatory methods to improve oxygenation, the prone positioning is leading with frequent and rapid increase in PaO2/FiO2 [2]. The combination of these conditions (prone and almitrine) could then be seen as a combination of gravitational and pharmacological effects to improve the intrapulmonary VA/Q mismatch. The increase of PaO2/FiO2 ratio in 80% of the patients confirmed the validity of this approach. The PaO2 increase associated with almitrine infusion was concomitant with a significant increase in ScvO2. This observation confirms the absence of tissue hypoperfusion with no large peripheral O2 extraction, as suggested by the low lactate levels. This ScvO2 increase provides a greater reserve for O2 extraction in case of acute desaturation, and increased the level of dissolved O2, the diffusible form of oxygen to the tissues.
The small size of the series would increase bias. The absence of a spontaneous PaO2 improvement over 8 hours in matched controls reinforced the credence in an almitrine effect. In absence of a drug shortage, all of these patients would have received almitrine, and we were able to administer the drug for a longer period than 36 hours in only few patients. This precluded any conclusion about the potential benefit on mechanical ventilation duration and on the number of prone positioning. Following the same line of thinking, the almitrine test on arterial oxygenation cannot be proposed as either a prognostic test or a predicator of the prone position response.
In conclusion, in a case series of early hypoxemic COVID-19 pneumonia with acute respiratory failure, IV almitrine was associated with an improvement in arterial blood oxygenation both in prone or supine positioning in most patients, suggesting a partial recovery of the pulmonary vessels’ contractility. This pharmacological intervention may offer an alternative and/or an additional strategy to the prone positioning in severe COVID-19 ARDS. It may help to support the lung function during a pandemic when the capacity to offer ECMO is very limited.
Contributions
MRL: supervision, data curation, methodology, validation, writing – original draft, writing – review & editing. Funding acquisition.
CL: data curation, visualisation.
MD: data curation, validation.
BC: visualisation, formal analysis, software, writing – review & editing.
DP: conceptualisation, formal analysis, visualisation, validation, methodology, writing – original draft, Writing – review & editing.
Funding
Université de Lorraine.
Disclosure of interest
The authors declare that they have no competing interest.
References
- 1.Gattinoni L., Coppola S., Cressoni M., Busana M., Rossi S., Chiumello D. Covid-19 does not lead to a “typical” acute respiratory distress syndrome. Am J Respir Crit Care Med. 2020:1299–1300. doi: 10.1164/rccm.202003-0817LE. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Munshi L., Del Sorbo L., Adhikari N.K.J., Hodgson C.L., Wunsch H., Meade M.O. Prone position for acute aespiratory distress syndrome. A systematic review and meta-analysis. Ann Am Thorac Soc. 2017;14:S280–S288. doi: 10.1513/AnnalsATS.201704-343OT. [DOI] [PubMed] [Google Scholar]
- 3.Payen D.M., Gatecel C., Plaisance P. Almitrine effect on nitric oxide inhalation in adult respiratory distress syndrome. Lancet. 1993;341:1664. doi: 10.1016/0140-6736(93)90801-m. [DOI] [PubMed] [Google Scholar]
- 4.B’Chir A., Mebazaa A., Losser M.R., Romieu M., Payen D. Intravenous almitrine bismesylate reversibly induces lactic acidosis and hepatic dysfunction in patients with acute lung injury. Anesthesiology. 1998;89:823–830. doi: 10.1097/00000542-199810000-00005. [DOI] [PubMed] [Google Scholar]
- 5.Reyes A., Roca J., Rodriguez-Roisin R., Torres A., Ussetti P., Wagner P.D. Effect of almitrine on ventilation-perfusion distribution in adult respiratory distress syndrome. Am Rev Respir Dis. 1988;137:1062–1067. doi: 10.1164/ajrccm/137.5.1062. [DOI] [PubMed] [Google Scholar]