Introduction and background

Cardiac arrest is one of the most common causes of death. Cardiac arrest is treated by cardiopulmonary resuscitation (CPR): in the case of out-of-hospital cardiac arrest (OHCA), the basic life support, consisting of ventilation and chest compression, should be carried out by medical laypersons immediately after the need for CPR has been established. The medical team arriving usually later continues the CPR with measures of advanced cardiac life support (ACLS). Advanced airway management (AAM) is part of the standard treatment during ACLS [1].

If the first attempt at intubation is successful during AAM, it is called first-pass intubation success (FPS). Current studies underline the importance of FPS during in-hospital ACLS regarding further resuscitation measures [2]. A failed initial intubation attempt during OHCA is an independent risk factor for decreased effectiveness of the whole ACLS [3]. This might be explained by reduced hands-on chest time, which is known to reduce chances for return of spontaneous circulation (ROSC) [1].

The primary aim of this study was to investigate the association between prehospital FPS and ROSC and the occurrence of ROSC and time to ROSC in OHCA patients. The secondary aim was to identify factors associated with FPS and ROSC.

Methods

Study design and setting

This was a retrospective multicenter registry study in five different physician-staffed ambulance stations in Germany where AAM is provided by physicians trained in emergency medicine. In this setting, all steps of CPR are performed according to the guidelines of the European Resuscitation Council [4]. This study was approved by the ethics committees of the University Hospitals of Jena (No. 2019-1318-Daten) and Tübingen (No. 116/2019BO2).

Participants

The inclusion criteria for the patients in the study were non-traumatic OHCA patients aged ≥ 18 years, on whom AAM was provided by prehospital emergency physicians in the German physician-staffed ambulance stations Jena, Meiningen, Tübingen, Wittlich and Wolfenbüttel.

Data collection and processing

Data were collected during an 18-month period from 1 July 2017 until 31 December 2018, utilizing the local intubation registry, the German Resuscitation Registry, and the emergency medical protocols. All registry data were gathered by online questionnaires.

The intubation registry includes information on AAM (performer, equipment, laryngoscopy, intubation method, medications, verification of tube position, respiratory evaluation, complications) and FPS. The German Resuscitation Registry includes information on CPR (initial heart rhythm, initial breathing, medications, defibrillation) and ROSC.

Result measurements

The main outcome variables were FPS, ROSC and time to ROSC. Because time to ROSC is not included as variable in the registry, it was calculated by taking the time difference between starting point of CPR and time of ROSC.

Statistical methods

Continuous variables were presented as mean ± standard deviation (SD) or median and interquartile ranges (IQR), categorical variables as number and percent.

To compare nominal variables with more than 2 categories in univariate analysis we used the χ2-test, for continuous variables the Mann-Whitney U test and for binary variables the Fisher’s exact test. Only significant variables were included in the multivariate binary logistic regression and odds ratios (OR) with 95% confidence interval (CI) of the model are reported. All statistical analyses were conducted using the IBM SPSS Statistics software (IBM Corp., Armonk, NY, USA; version 25), the significance level was set at 0.05 for each test.

Results

Patient characteristics

During the 18-month study period, 213 patients were recorded in the intubation registry, of which 180 patients were included in the study and analyzed, with most of the exclusions caused by incomplete questionnaires. The study population showed the following demographics: average age of 69.7 years (SD 12.9 years), mean age of 70 years (IQR 60–79.75 years), median body weight of 82.5 kg (IQR 75–95kg) and 71.7% male patients.

An FPS was recorded in 150 patients (83.3%), and ROSC was achieved in 82 patients (45.5%).

Association between FPS and ROSC and time to ROSC

There was a positive association between FPS and ROSC (p = 0.027, Table 1). In multivariate analysis, FPS was the only factor independently associated with ROSC, with an OR of 5.281 (p = 0.002, 95% CI 1.800–15.494) (Table 2). In patients with FPS, a statistically non-significant trend for shorter time to ROSC was observed: median 18min (IQR 7–27 min) vs. median 28 min (IQR 21–35 min) (p = 0.059, Table 3).

Table 1 Comparisons between the groups with and without prehospital ROSC
Table 2 Multivariate binary logistic regressions for FPS and ROSC
Table 3 Comparisons between the groups with and without FPS

Factors associated with FPS

Patients were divided into two groups based on the presence of prehospital FPS (FPS, n = 150, 83.3%; No FPS, n = 30, 16.7%, Table 3).

In univariate analysis (Table 1), the following factors were associated with FPS: body weight (p = 0.044), throat suction (p = 0.001), verification of tube position (p < 0.001), airway assessment according to Cormack and Lehane (p < 0.001) and complications during intubation (p < 0.001). The following factors were not associated with FPS: age (p = 0.186), sex (p = 0.182), level of education of performer (p = 0.180), specialization of performer (p = 0.335), selected intubation equipment (p = 0.197), type of laryngoscopy (p = 0.062), intubation method (p = 0.901), specific medications (analgesics, p = 0.130; sedative, p = 0.692; muscle relaxants, p = 0.585) and time to ROSC (p = 0.059).

In multivariate analysis (Table 2), the use of capnography (OR = 7.384, 95% CI: 1.886–28.917) and complications during AAM (OR = 0.033, 95% CI: 0.007–0.153) were independently associated with FPS.

Factors associated with ROSC

Patients were divided into two groups based on the presence of prehospital ROSC (ROSC, n = 82, 45.6%; No ROSC, n = 98, 54.4%; Table 1). The average time to ROSC was 22.16 min (SD 12.7 min), the mean time was 20 min (IQR 14.5–28.5 min).

In univariate analysis (Table 1), the following factors were associated with ROSC: initial hearth rhythm (p = 0.001), initial breathing (p = 0.001), amiodarone (p = 0.037) and FPS (p = 0.027). The following factors were not associated with ROSC: age (p = 0.148), sex (p = 0.246), body weight (p = 0.057) and number of defibrillations (p = 0.227).

In multivariate analysis (Table 2), only FPS was independently associated with ROSC (OR = 5.281, 95% CI: 1.800–15.494).

Discussion

This study showed that in adult OHCA patients with AAM prehospital FPS was associated with a higher chance of ROSC. Factors associated with FPS positively were the use of capnography for confirmation of tube position, and no complications during AAM. Only FPS was a factor associated with ROSC.

In recent studies, FPS has been in the spotlight, e.g. a systematic review examined the emergency AAM in general and concluded that preclinical physicians achieved an FPS rate of 71.2–87.5% [2]. The FPS rate in this study (83.3%) during prehospital CPR is in line with these results. Since there was no connection between the FPS and the level of education of performer and specialization of performer in this study, one can assume good intubation expertise on the part of the emergency physicians for the data examined. Compared to pan-European data (32.7%) [9] the rate of prehospital ROSC (45.6%) was higher. This might be explained by the fact that in this study AAM was mostly provided by emergency physicians and not by paramedics.

Regarding FPS, the negative association of body weight to FPS can be explained by AAM itself becoming more difficult [10]. Using capnography for confirmation of correct tube position is required in the current resuscitation guidelines as the treatment standard [1]. The results of this study support this fact, because using capnography was independently associated with higher chance of FPS. The available data also show that the application rate of capnography was not 100%. This is probably due to the fact that the questionnaire differentiates between capnography and capnometry and is marginally misleading when it is filled out. This fuzziness should be examined more profoundly in the future.

Over the past few years, many studies have been carried out researching the best equipment for AAM [5,6,7,8]. This study cannot draw any sound conclusions about the best equipment, because ETI is overpresented (85.3% of all intubations).

Regarding complications during AAM (complication rate 14.2%), the most common documented was failed intubation followed by aspiration. Statistically, in this study complications during AAM were independently associated with FPS. Here, future studies should clarify the connection, because it is not clear whether the FPS failed due to complications during the AAM or whether an FPS was present before the emergency physicians noticed those complications (e.g. later aspiration). Further improvement of the questionnaires to depict the chronological sequence of event should occur.

The findings of this study show that a shockable rhythm and present gasping have a positive effect on probability of ROSC by intervening in the circulatory arrest in an earlier more favorable phase, which is consistent with other studies [3, 11, 12]. Compared to other studies, time to ROSC is slightly longer, which can be justified by methodological differences on calculating time to ROSC [11, 13]. Therefore, it would be desirable to integrate the variable time to ROSC into the resuscitation registers in order to ensure clear comparability.

The results confirm the significant positive association between prehospital FPS and ROSC [3, 14], with the distinctiveness that this study found this association in a setting with physician-staffed ambulance stations. If FPS was achieved, time to ROSC was lowered by an average of 10 min [3].

As prolonged intubation attempts usually occur while stopping chest compressions, an extended no-flow time might reduce the chance to achieve a ROSC and therefore affirm the importance of a fast FPS. Other reasons for the lower rate of ROSC might be a delay of other necessary interventions such as defibrillation or detecting and treating reversible causes of the cardiac arrest.

In future this study design should be repeated with larger power to see whether there is a statistically significant difference in time to ROSC depending on prehospital FPS.

Additionally, the length of the interruptions for intubating (if present) while performing chest compressions and subsequent disturbances if the ACLS algorithm should be further queried in the questionnaires. To this end, the questionnaires could be modified so that the medical staff could assess the effectiveness of their chest compressions and the amount of time they spent on resuscitation on a scale.

Limitations

This study has several limitations. First, the results are based on a retrospective analysis of 2 registries with the possible bias by the intrinsic design itself.

Second, the full treatment time was not always documented (start and end), therefore valid statements about survival times could not be made.

Third, this study only looked at appearance of preclinical ROSC. It is not possible to draw any conclusions as to whether the patients achieved a ROSC later in the hospital and no evaluation of neurological outcome.

Fourth, the informative value is limited in that the group sizes “FPS” and “No FPS” differ greatly from one another in quantitative terms. Attention should be paid to greater power in future investigations and possibly more locations to be included.

Conclusion

These results indicate that in adult OHCA patients with AAM, prehospital FPS is associated with a higher chance of ROSC. Additionally, the use of capnography and no complications during AAM were positive associated with the successful first-pass attempt.