If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pathophysiology and Transplantation, University of Milan, ItalyDepartment of Anesthesiology, Intensive Care and Emergency, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy
Department of Anaesthesiology, Intensive Care and Pain Medicine, University of Helsinki and Helsinki University Hospital, Helsinki, FinlandDepartment of Emergency Care and Services, Helsinki University Hospital and University of Helsinki, Helsinki, Finland
We compared the prognostic abilities of neurofilament light (NfL) and neuron-specific enolase (NSE) in patients resuscitated from out-of-hospital cardiac arrest (OHCA) of various aetiologies.
Methods
We analysed frozen blood samples obtained at 24 and 48 hours from OHCA patients treated in 21 Finnish intensive care units in 2010 and 2011. We defined unfavourable outcome as Cerebral Performance Category (CPC) 3–5 at 12 months after OHCA. We evaluated the prognostic ability of the biomarkers by calculating the area under the receiver operating characteristic curves (AUROCs [95% confidence intervals]) and compared these with a bootstrap method.
Results
Out of 248 adult patients, 12-month outcome was unfavourable in 120 (48.4%). The median (interquartile range) NfL concentrations for patients with unfavourable and those with favourable outcome, respectively, were 689 (146–1804) pg/mL vs. 31 (17–61) pg/mL at 24 h and 1162 (147–4360) pg/mL vs. 36 (21–87) pg/mL at 48 h, p < 0.001 for both. The corresponding NSE concentrations were 13.3 (7.2–27.3) µg/L vs. 8.5 (5.8–13.2) µg/L at 24 h and 20.4 (8.1–56.6) µg/L vs. 8.2 (5.9–12.1) µg/L at 48 h, p < 0.001 for both. The AUROCs to predict an unfavourable outcome were 0.90 (0.86–0.94) for NfL vs. 0.65 (0.58–0.72) for NSE at 24 h, p < 0.001 and 0.88 (0.83–0.93) for NfL and 0.73 (0.66–0.81) for NSE at 48 h, p < 0.001.
Conclusion
Compared to NSE, NfL demonstrated superior accuracy in predicting long-term unfavourable outcome after OHCA.
Prognostication after cardiac arrest (CA) should be performed using a multimodal approach, including clinical assessment, neurophysiology, radiological investigations and biomarkers.
The updated European Resuscitation Council (ERC)-European Society of Intensive Care Medicine (ESICM) guidelines recommend using neuron-specific enolase (NSE) as one component of multimodal prognostication.
However, the recommended high NSE cut-off values that are necessary to achieve high specificity may result in low sensitivity in detecting patients with poor prognosis.
One example is the decreased prognostic accuracy in elderly patients and patients with a short time from collapse to return of spontaneous circulation (ROSC).
Comparison of three analytical platforms for quantification of the neurofilament light chain in blood samples: ELISA, electrochemiluminescence immunoassay and Simoa.
NfL demonstrated a very high capacity to predict unfavourable six-month outcome after out-of-hospital cardiac arrest (OHCA) with a presumed cardiac cause.
NfL also appeared to have the best ability among a group of neurobiomarkers, including NSE, to find patients with a favourable outcome despite the indeterminate prognosis given by examinations recommended in the ERC-ESICM guidelines.
Before wider adoption, the utility and presumed superiority of NfL over NSE should be validated also in unselected CA populations. Accordingly, we analysed NfL concentrations and its prognostic capacity in an unselected OHCA population, including patients with shockable and non-shockable initial rhythms and resuscitated from different CA aetiologies. We hypothesised that NfL would be superior to NSE in predicting unfavourable long-term outcome in patients treated in the intensive care unit (ICU) following OHCA. The secondary hypothesis was that NfL would have better prognostic value in those patient subgroups (high age, short time from collapse to ROSC) where NSE has demonstrated poor prognostic accuracy.
Methods
Study population and definitions
This was a post hoc analysis of the prospective multicentre FINNRESUSCI study of 548 adult patients resuscitated after OHCA and treated in 21 Finnish ICUs between 2010 and 2011.
All five university hospitals and 14 out of 15 non-university central hospitals participated in FINNRESUSCI. Over 98% of the Finnish population live in the referral areas of these hospitals. The FINNRESUSCI study protocol was approved by the Helsinki University Hospital Ethics Committee and by each participating hospital. A post-hoc substudy of NSE values was published earlier.
In this post-hoc study, we included 248 patients whose blood samples were stored (Fig. 1). We defined outcome according to the Cerebral Performance Category (CPC)
at 12 months after CA: CPC 1–2 indicates favourable outcome and CPC 3–5 indicates unfavourable outcome. The CA cause was defined with clinical criteria.
Fig. 1Flowchart of the study population. Abbreviations: NfL: neurofilament light. NSE: neuron-specific enolase.
The patient data were collected using Internet-based forms. Data on previous health status were collected from the patients’ medical history and mortality data from Statistics Finland. The outcome according to CPC classification was assessed 12 months after CA with phone interviews conducted by a neurology specialist blinded to the hospital treatment and the laboratory analysis.
Blood samples
The blood samples were from patients in the FINNRESUSCI study for whom the next of kin had provided written informed consent. The plasma samples were collected at 24 and 48 hours from OHCA, stored at −80 °C and thawed for this analysis. We measured the NfL levels quantitatively using a commercially available two-step digital immunoassay using the single molecule array Quanterix SIMOA™ NF-light® Kit and SIMOA™ HD-1 Analyzer (SIMOA™, Quanterix Corporation, Lexington, MA, USA). The plasma NfL concentrations were expressed in picograms per millilitre (pg/mL). For comparison, we used NSE samples from the same time points determined according to previously described methods.
The obtained serum samples were stored at −70 °C during the original study and analysed with a commercially available electrochemiluminescence immunoassay (Roche Diagnostics GmbH, Mannheim, Germany) in April 2015. We excluded all NSE samples with significant haemolysis, ≥ 500 mg/L.
TTM-Trial Investigators. Neuron-specific enolase as a predictor of death or poor neurological outcome after out-of-hospital cardiac arrest and targeted temperature management at 33°C and 36°C.
We present categorical data as absolute numbers with percentages (95% confidence intervals [CIs]) and continuous data as medians with interquartile ranges (IQRs). For continuous data, we used Student’s t test (normal distribution) and the Mann-Whitney U test or the Kruskal-Wallis test (skewed distribution) for comparison. We compared the categorical variables with the Chi square test or Fisher’s exact test. We divided the study population into quartiles according to patients’ age and time to ROSC
to detect differences in prognostic values between NfL and NSE.
We calculated the areas under the receiver operating characteristic curves (AUROCs) with 95% CIs to assess the ability of NfL and NSE to discriminate between patients in favourable (CPC 1–2) and those in unfavourable (CPC 3–5) outcome groups. We compared the AUROCs of NfL to NSE at 24 and 48 h after CA using the bootstrap method. We constructed a multivariable model with clinical factors such as age, initial rhythm, delay to ROSC and witnessed collapse for the prediction of poor functional outcome. Into this model, with a backward stepwise approach, we subsequently inserted NfL and NSE and report results with odds ratios and 95% CIs.
We defined the NfL cut-off values to predict unfavourable outcome at 24 and 48 h after CA from the receiver operating characteristic curve and for NSE at 48 h, accordingly. The cut-off values for NSE at 24 h were not calculated because of its poor prognostic accuracy.
We determined biomarker concentrations for high specificity (low false positive rate, [FPR]) to detect patients with a high probability for unfavourable outcome and concentrations for high sensitivity to detect those with a high probability for favourable outcome (low false negative rate). We calculated the Youden-based
cut-off values to assess the concentrations that simultaneously have as high specificity and sensitivity as possible, to promote their comparability. Furthermore, we defined cut-offs for high sensitivity (95% and 99%) and used normal levels of NfL to detect patients with favourable outcome. We used concentrations of 55 pg/mL for NfL
as the highest normal value. We also calculated the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+) or negative likelihood ratio (LR-), if suitable, for these cut-off values. We considered p values < 0.05 as significant. We performed statistical analyses with SPSS version 27 (SPSS, Il, Chicago, USA) and R program, version 4.0.0.
Results
The 12-month outcome was unfavourable in 120/248 (48.4%) of the patients. Of these patients, 177 (71.4%) had a shockable initial rhythm. Blood samples enabled NfL analysis in 243 patients and NSE analysis in 248 patients (Fig. 1). Table 1 shows the outcome data and patient characteristics. The comparison of the study patients to the FINNRESUSCI patients in whom blood samples were unavailable are shown in Table S1.
Table 1Characteristics of the study patients according to Cerebral Performance Category classification.
The NfL concentrations were significantly higher for the patients with unfavourable outcome than for those with favourable outcome at all time points. At 24 h, the median concentrations (IQR) were 688.9 pg/mL (146.1–1803.8) for the patients with unfavourable outcome vs. 30.9 pg/mL (16.9–61.2) pg/mL for those with favourable outcome (p < 0.001). Accordingly, the concentrations at 48 h were 1162.4 pg/mL (146.8–4360.5) vs. 35.6 pg/mL (21.3–86.7), p < 0.001. Fig. 2 shows the concentrations indexed by outcome. The NSE concentrations were higher for the patients with unfavourable outcome than for those with favourable outcome; at 24 h, the concentrations were 13.3 µg/L (7.2–27.3) for the patients with unfavourable outcome vs. 8.5 µg/L (5.8–13.2) for those with favourable outcome, p < 0.001. At 48 h, the concentrations were 20.4 µg/L (8.1–56.6) vs. 8.2 µg/L (5.9–12.1), respectively (p < 0.001) (Fig. 2). The NfL and NSE concentrations were not different for the patients with a cardiac aetiology of arrest compared to those with a non-cardiac aetiology, according to outcome Table S2.
Fig. 2Boxplots for NfL (A) and NSE (B) concentrations at 24 h and 48 h after cardiac arrest for patients with favourable (CPC 1–2) and unfavourable (CPC 3–5) outcomes with a 10-based logarithmic scale. Each box presents the interquartile range. The line inside the box shows the median value, the whiskers show the lowest and the highest concentrations, and the dots show the concentrations for each individual. Abbreviations: CPC: Cerebral Performance Category. NfL: neurofilament light. NSE: neuron-specific enolase.
The prognostic ability assessed with AUROC (with 95% CI) was significantly higher at 24 h after CA to predict unfavourable outcome for NfL (0.90 [0.86–0.94]) than NSE (0.65 [0.58–0.72]), p < 0.001. At 48 h, the AUROC was higher for NfL (0.88 [0.83–0.94]) than NSE (0.72 [0.66–0.81]), p < 0.001. The AUROC for NfL at 24 h was also higher than NSE at 48 h, p < 0.001. NfL at 24 h was a significant predictor of unfavourable outcome in the multivariable model, whereas NSE at 48 h was not (Table S3). The AUROCs for NfL and NSE according to the CA aetiology are presented in Table S2.
Cut-off values
The NfL cut-off values to predict unfavourable outcome using the Youden method (maximising sensitivity and specificity) were 97 pg/mL at 24 h and 231 pg/mL at 48 h. For those cut-offs, the specificities (with 95% CIs) were 86.8% (80.6–93.0) and 92.1% (86.8–97.3), and the sensitivities were 81.8% (74.2–88.6) and 72.2 (62.3–82.0), respectively. For 99% specificity, the cut-offs were 589 pg/mL and 721 pg/mL, respectively with sensitivities of 54.0% (44.8–63.2) and 59.5% (48.7–70.3), respectively (Table 2).
Table 2Characteristics (with 95% CIs) of cut-off values for NfL at 24 h and 48 h and for NSE at 48 h after cardiac arrest for high demand of specificities to predict unfavourable outcome.
Regarding NSE at 48 h, using a 35 µg/L cut-off value with 99% specificity resulted in a 37.1% (27.5–46.7) sensitivity. Table 2 shows the cut-off values with corresponding characteristics for NfL at 24 h and 48 h and for NSE at 48 h using the Youden method and 95% and 99% specificity. The cut-off values for NfL and NSE to predict favourable outcome are presented in Table S4.
Combining NfL and NSE, 0.3% of the patients who exceeded the cut-offs for 95% specificity had a favourable outcome (Table S5).
NfL in different subgroups
Age quartiles
In all age groups, the NfL concentrations were significantly higher for the patients with unfavourable outcome than for those with favourable outcome at 24 h and 48 h after CA (Fig. 3). The prognostic ability of NfL was significantly better at 24 h than that of NSE in all age subgroups (Table S6). At 48 h, the prognostic ability of NfL was better than that of NSE in the patients aged 57–63 years (p = 0.005) and in the oldest subgroup, ≥72 years (p = 0.020) (Table S6). The AUROC for NfL to predict unfavourable outcome was lower in the oldest quartile compared to the youngest quartile (18–56 years) both at 24 h (p = 0.016) and 48 h (p = 0.032). The AUROC was also lower in the fourth quartile (≥72 years) at 48 h than in the second quartile (57–63 years), p = 0.020. The NfL concentrations in the patients with favourable outcome were significantly different according to age group at 24 h (p < 0.001) and at 48 h (p = 0.001).
Fig. 3Boxplots for NfL concentrations at 24 h and 48 h after cardiac arrest for patients with favourable (CPC 1–2) and unfavourable (CPC 3–5) outcomes with a 10-based logarithmic scale, according to different age quartiles. Each box presents the interquartile range. The line inside the box shows the median value, the whiskers show the lowest and the highest concentrations, and the dots show the concentrations for each individual. Age intervals (years) with p values (for differences in concentrations for patients with favourable [CPC 1–2] and unfavourable [CPC 3–5] outcomes in each quartile) are presented above each figure. Abbreviations: CPC: Cerebral Performance Category. NfL: neurofilament light.
The NfL concentrations were significantly higher for the patients with unfavourable outcome compared to those with favourable outcome at all times from collapse to ROSC quartiles at 24 h and 48 h (Fig. 4). The prognostic ability of NfL was also better than that of NSE in all ROSC subgroups at 24 h and 48 h after CA (Table S7). The AUROC for NfL to predict unfavourable outcome was lower in the quartile with the shortest time from collapse to ROSC (1–13 min) than in the quartile with the longest time to ROSC (≥29 min) at 24 h (p = 0.014) and at 48 h (p = 0.019). The AUROC was also lower for the patients in the second quartile (ROSC 14–20 min) than for those in the fourth quartile (≥29 min) at 48 h, p = 0.032. The distributions of NfL concentrations were significantly different according to outcome in the ROSC subgroups: for the patients with favourable outcome (at 24 h p = 0.034; at 48 h p = 0.004) and for those with an unfavourable outcome (p < 0.001 at 24 h and 48 h).
Fig. 4Boxplots for NfL concentrations at 24 h and 48 h after cardiac arrest for patients with favourable (CPC 1–2) and unfavourable (CPC 3–5) outcomes with a 10-based logarithmic scale, according to different ROSC quartiles. Each box presents the interquartile range. The line inside the box shows the median value, the whiskers show the lowest and the highest concentrations, and the dots show the concentrations for each individual. ROSC intervals (minutes) with p values (for differences in concentrations for patients with favourable [CPC 1–2] and unfavourable [CPC 3–5] outcomes in each quartile) are presented above each figure. Abbreviations: CPC: Cerebral Performance Category. NfL: neurofilament light. ROSC: return of spontaneous circulation.
In this post-hoc analysis of OHCA patients resuscitated from various arrest aetiologies, NfL was significantly more accurate than NSE in predicting unfavourable 12-month outcome. The prognostic ability of NfL was already excellent at 24 hours after CA. The median concentrations for the patients with unfavourable outcome were about 20-fold greater than for those with favourable outcome. Importantly, NfL was also accurate in the patients resuscitated from a likely non-cardiac cause of arrest. We also found a less clear association between age and time to ROSC and predictive accuracy than we previously showed with NSE.
As our sample presents heterogeneous OHCA patients, our findings support wider utilisation of NfL in clinical prognostication after CA.
The lack of wider adoption of NfL thus far may have been due to the unavailability of a commercial assay, but given the introduction of the ultrasensitive SIMOA method, this is likely to change.
However, few studies exist about prognostication after CA using the ultrasensitive SIMOA method. NfL measurement within the first 24 h after ROSC demonstrated an AUROC of 0.82 to predict in-hospital death.
In a Targeted Temperature Management (TTM) substudy including 782 OHCA patients with a likely cardiac aetiology of arrest, the AUROCs at 24–72 h to predict poor six-month outcome were 0.94–0.95.
The present study, including an unselected population with both shockable and non-shockable rhythms, found AUROCs to predict CPC 3–5 at 12 months of 0.88–0.90, demonstrating slightly worse but still excellent discriminative ability. Pouplet et al demonstrated AUROC of 0.87 to predict CPC 3–5 at 90 days after CA in patients with shockable rhythms using different but comparable commercial laboratory method.
Pouplet C, Colin G, Guichard E, et al. AfterROSC network. The accuracy of various neuro-prognostication algorithms and the added value of neurofilament light chain dosage for patients resuscitated from shockable cardiac arrest: an ancillary analysis of the ISOCRATE study. Resuscitation 2021:S0300-9572(21)00517-7. https://doi.org/10.1016/j.resuscitation.2021.12.009. Epub ahead of print. PMID: 34915084.
TTM-Trial Investigators. Neuron-specific enolase as a predictor of death or poor neurological outcome after out-of-hospital cardiac arrest and targeted temperature management at 33°C and 36°C.
We found a slightly lower discriminative ability, especially for NSE, than previously reported. The likeliest explanation is the inclusion of different types of CA patients in whom the reason for the unfavourable outcome may not only be due to post-cardiac arrest brain injury (PCABI), which is the most common cause of death after CA.
Clearly, NfL and NSE can only work for predicting death or poor outcome related to brain injury.
The levels of NSE for patients with unfavourable outcome were somewhat lower in this study compared to some previous studies. There are several possible explanations for this. Firstly, the laboratory methods used may be important.
Secondly, it is possible that the lower levels and prognostic ability of NSE seen in the present study compared to previous studies are related to differences in the definition of unfavourable outcome
TTM-Trial Investigators. Neuron-specific enolase as a predictor of death or poor neurological outcome after out-of-hospital cardiac arrest and targeted temperature management at 33°C and 36°C.
Our secondary finding was that NfL’s prognostic ability was better than NSE in subgroups where the prognostic value of NSE was poor, such as the elderly and those with a shorter arrest duration. In our study, the NfL levels were higher in those with longer time from collapse to ROSC, and the accuracy was highest in those with the longest time to ROSC. However, even in the group with a short time to ROSC, the discriminative ability was satisfactory. This may suggest that NfL is more sensitive even in detecting milder hypoxic brain injury. Importantly, for patients with a short time from collapse to ROSC and patients aged ≥72 years, the prognostic value of NfL was superior to NSE. Increasing age is one confounding factor of NfL; the concentrations increase about 2% per year,
We also found a rising trend of NfL levels in CPC 1–2 patients with increasing age. This finding may provide an additional explanation for the worse discriminative ability of NfL in the oldest patient group.
The ERC-ESICM guidelines recommend a 60 µg/L NSE cut-off.
In this study, the 35 µg/L NSE cut-off at 48 h yielded 99% specificity but 37% sensitivity. Generally, demanding a very high specificity results in low sensitivity if the diagnostic method’s performance is insufficient.
Targeting specificities of 95% and 99%, the cut-off values for NfL at 24 h were 232 pg/mL and 589 pg/mL, respectively; the cut-off values for NfL at 48 h were 445 pg/mL and 721 pg/mL, respectively.
Those cut-off concentrations are comparable to corresponding values in a TTM substudy.
The Youden-based NfL cut-offs showed 72–82% sensitivities and 87–91% specificities. In this study population, NfL presented better sensitivity than NSE, even with clinically useful specificities. The combination of cut-offs of NfL and NSE for 95% specificity resulted in a 0.3% FPR.
Recent studies have raised the concern that there might be CA patients with potentially favourable outcome despite poor prognosis given by prognostic methods.
Late awakening, prognostic factors and long-term outcome in out-of-hospital cardiac arrest - results of the prospective Norwegian Cardio-Respiratory Arrest Study (NORCAST).
Targeting 95% and 99% specificity to find patients with favourable outcome, the NfL cut-offs were 14–29 pg/mL, which are in the normal range. NSE demonstrated insufficient capacity to detect patients with favourable outcome. NfL has a better ability than NSE to find patients with favourable outcome using normal or lower values.
Strengths and limitations
Our study has several strengths. It was a nationwide multicentre study with a large patient sample from many ICUs. Importantly, we included CA patients of various arrest aetiologies. The treating clinicians were blinded to the NfL results. Neurological outcome was defined by an experienced neurologist blinded to the biomarker results. However, some limitations exist. First, the original study is 10 years old, and prognostication and clinical care of resuscitated patients are likely to have changed. Second, our study population was selected by consent availability, and, consequently, the proportion of patients with bystander cardiopulmonary resuscitation, shockable rhythm and TTM was significantly higher in those included than those excluded. Third, we do not have conclusive data on the patients’ cause of death or prognostication; the patients were managed according to protocols available at the time. Fourth, the numbers of patients in the subgroups were small.
Conclusion
NfL is more valuable than NSE in prognostication of unfavourable outcome after OHCA, also in cases with non-cardiac aetiologies. Contrary to NSE, NfL retained its accuracy in the elderly and those with a short delay to ROSC, suggesting the ability of NfL to also identify milder forms of hypoxic brain injury.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
CRediT authorship contribution statement
L. Wihersaari: Writing – original draft, Formal analysis, Visualization, Software, Data curation. M. Reinikainen: Conceptualization, Methodology, Writing – original draft, Supervision. R. Furlan: Resources, Validation. A. Mandelli: Resources, Validation. J. Vaahersalo: Validation, Investigation. J. Kurola: Validation, Conceptualization. M. Tiainen: Validation, Investigation. V. Pettilä: Validation, Supervision. S. Bendel: Validation, Supervision. T. Varpula: Validation, Conceptualization. R. Latini: Resources, Validation. G. Ristagno: Resources, Validation. MB. Skrifvars: Writing – original draft, Conceptualization, Methodology, Project administration, Supervision.
Acknowledgements
The manuscript writing was funded by a grant from Finska Läkaresällskapet. The authors thank biostatistician Tuomas Selander, MSc, for helping with the statistical analyses.
The FINNRESUSCI Study Group: Satakunta Central Hospital, Dr. Vesa Lund, Päivi Tuominen, Satu Johansson, Pauliina Perkola, Elina Kumpulainen; East Savo Central Hospital, Dr. Markku Suvela, Sari Hirvonen, Sirpa Kauppinen; Central Finland Central Hospital, Dr. Raili Laru-Sompa, Mikko Reilama; South Savo Central Hospital, Dr. Heikki Laine, Pekka Kettunen, Iina Smolander; North Karelia Central Hospital, Dr. Matti Reinikainen, Tero Surakka; Seinäjoki Central Hospital, Dr. Kari Saarinen, Pauliina Lähdeaho, Johanna Soini; South Carelia Central Hospital, Dr. Seppo Hovilehto; Päijät-Häme Central Hospital, Dr. Pekka Loisa, Alli Parviainen, Pirjo Tuomi; Vaasa Central Hospital, Dr. Simo-Pekka Koivisto, Dr. Raku Hautamäki; Kanta-Häme Central Hospital, Dr. Ari Alaspää, Tarja Heikkilä; Lappi Central Hospital, Dr. Outi Kiviniemi, Esa Lintula; Keski-Pohjanmaa Central Hospital, Dr. Tadeusz Kaminski, Jane Roiko; Kymenlaakso Central Hospital, Dr. Seija Alila, Dr. Jussi Pentti, Reija Koskinen; Länsi-Pohja Central Hospital, Dr. Jorma Heikkinen; Helsinki University Hospital, Jorvi Hospital, Dr. Jukka Vaahersalo, Dr. Tuomas Oksanen, Dr. Tero Varpula, Anne Eronen, Teemu Hult, Taina Nieminen; Meilahti Hospital Medical ICU, Dr. Tom Bäcklund, Leevi Kauhanen; Meilahti Hospital ICU, Dr. Kirsi-Maija Kaukonen, Dr. Ville Pettilä, Leena Pettilä, Sari Sutinen; Turku University Hospital, Dr. Juha Perttilä, Keijo Leivo; Tampere University Hospital, Dr. Sanna Hoppu, Dr. Jyrki Tenhunen, Dr. Sari Karlsson, Atte Kukkurainen, Simo Varila, Samuli Kortelainen, Minna-Liisa Peltola; Kuopio University Hospital, Dr. Pamela Hiltunen, Dr. Jouni Kurola, Dr. Esko Ruokonen, Elina Halonen, Saija Rissanen, Sari Rahikainen; Oulu University Hospital, Dr. Risto Ahola, Dr. Tero Ala-Kokko, Sinikka Sälkiö.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Comparison of three analytical platforms for quantification of the neurofilament light chain in blood samples: ELISA, electrochemiluminescence immunoassay and Simoa.
TTM-Trial Investigators. Neuron-specific enolase as a predictor of death or poor neurological outcome after out-of-hospital cardiac arrest and targeted temperature management at 33°C and 36°C.
Pouplet C, Colin G, Guichard E, et al. AfterROSC network. The accuracy of various neuro-prognostication algorithms and the added value of neurofilament light chain dosage for patients resuscitated from shockable cardiac arrest: an ancillary analysis of the ISOCRATE study. Resuscitation 2021:S0300-9572(21)00517-7. https://doi.org/10.1016/j.resuscitation.2021.12.009. Epub ahead of print. PMID: 34915084.
Late awakening, prognostic factors and long-term outcome in out-of-hospital cardiac arrest - results of the prospective Norwegian Cardio-Respiratory Arrest Study (NORCAST).
☆Some results were presented as abstracts at the 34th ESICM LIVES Digital Annual Congress, 3–6 October 2021 and at Operatiiviset Päivät 17–19 November 2021, Helsinki, Finland.
00063-6&id=gr1.jpg){kind=link}
00063-6&id=gr2.jpg){kind=link}
00063-6&id=gr3.jpg){kind=link}
00063-6&id=gr4.jpg){kind=link}