ABSTRACT
INTRODUCTION
Miller Fisher syndrome (MFS) is typically associated with a benign course and spontaneous recovery. However, immunotherapies such as intravenous immunoglobulin (IVIG) and plasma exchange (PE) are often administered for this condition because of the potential for clinical deterioration. We aimed to evaluate the real-world use of immunotherapies in MFS management using a nationwide Japanese database.
METHODS
We conducted a retrospective cohort study between April 2014 and March 2020 using the Japanese Diagnosis Procedure Combination database. Patient demographics and treatment modalities were described. Outcomes included length of hospital stay, in-hospital mortality, hospitalization cost, and activities of daily living defined by the Barthel index score.
RESULTS
We identified 1,595 patients with MFS. Of these, 999 (62.6%) received immunotherapy (including IVIG for 908 [56.9%] patients, intravenous methylprednisolone for 219 [13.7%] patients, and PE for 18 [1.8%] patients). In patients with and without immunotherapy, the median length of hospital stay was 18 and 12 days, in-hospital mortality was 0.7% and 0.2%, median total hospitalization costs were ¥1,660,200 and ¥549,375, and the proportions of Barthel index score of 95–100 at discharge were 75.3% and 75.2%, respectively.
CONCLUSION
Our findings highlight the widespread use of immunotherapies for MFS in Japan despite its generally favorable natural course.
INTRODUCTION
Miller Fisher syndrome (MFS) is a rare subgroup of Guillain–Barré syndrome (GBS), an acute immune-mediated neuropathy. MFS is characterized by the clinical triad of acute external ophthalmoplegia, ataxia, and areflexia1),2), which typically develop after upper respiratory tract infections. MFS symptoms generally progress over 1–2 weeks and often improve spontaneously. However, in some cases, MFS evolves into a severe condition, such as Bickerstaff’s brainstem encephalitis , which is characterized by impaired consciousness and cranial nerve palsy. Owing to the shared clinical features and the presence of anti-ganglioside antibodies, particularly the anti-GQ1b antibody, MFS is closely related to both GBS and Bickerstaff’s brainstem encephalitis.
Immunotherapies, such as intravenous immunoglobulin (IVIG) and plasma exchange (PE), are typically used to treat the acute phases of GBS and Bickerstaff’s brainstem encephalitis3). However, for MFS, immunotherapy is recommended when the condition of the patient deteriorates4). Although several studies have attempted to identify predictors of worsening MFS5),6), standardized criteria for the initiation of immunotherapy remain elusive. Consequently, treatment decisions are predominantly made at the discretion of clinicians.
Because immunotherapies are associated with high costs and potential risks, their appropriate use is critical for optimal clinical care and resource management7). In this study, we used a national inpatient database in Japan to provide an overview of the use of immunotherapy for managing patients diagnosed with MFS. By assessing the actual status of MFS treatment in the real world, we identified the factors that need to be addressed to promote appropriate treatment decisions.
METHODS
In this retrospective cohort study, we used data from the Japanese Diagnosis Procedure Combination database8). The database includes inpatient administrative and discharge abstract claims from >1,200 hospitals and contains data on >50% of hospitalized patients who have undergone acute care9). The database includes the main diagnoses, comorbidities present at admission, and comorbidities occurring after admission for each patient, which are recorded using the International Classification of Diseases, 10th Revision codes (ICD-10), and text data. The database also contains the following patient information: age, sex, height, weight, smoking status, Barthel index score, prescription and treatment records, discharge status, and hospital code. The diagnostic records in the database were well validated, and the sensitivity and specificity of the primary diagnoses were 50–80% and 96%, respectively10).
We identified all patients with a main diagnosis at discharge of “Miller-Fisher syndrome” or “Fisher syndrome” (ICD-10: G61.0 with Japanese text of diagnosis) between April 2014 and March 2020. We collected data on the use of immunotherapy (IVIG, PE, and intravenous methylprednisolone).
We categorized the patients into the following age groups: ≤19, 20–39, 40–59, 60–79, and ≥80 years. The Japan Coma Scale, which correlates well with the Glasgow Coma Scale, was categorized as 0 (fully alert), 1-digit (awake without any stimuli), 2-digit (aroused by some stimuli), and 3-digit (not aroused by any stimuli)11),12). Activities of daily living (ADL) at the time of admission were recorded using the Barthel index, in which a higher score indicated preserved functional capacity. The Barthel index was categorized into four score groups: 95–100, 65–90, 25–60, and 0–20. Admissions were classified as planned or unplanned. Hospitals were categorized as academic (university hospitals) or non-academic (all other hospitals) based on their institutional codes.
The following outcomes were evaluated in the immunotherapy and non-immunotherapy groups: length of hospital stay, in-hospital death, total hospitalization costs, Japan Coma Scale score at discharge, Barthel index score at discharge, and difference in the Barthel index score between the time of admission and discharge. To assess the variability in treatment decisions across facilities, we calculated the proportion of patients receiving immunotherapy in each hospital with five or more cases of MFS during the study period.
Categorical variables are shown as numbers and percentages and were compared using Fisher’s exact test. Continuous variables were compared using Student’s t-test or the Mann–Whitney U test and are represented as the mean and standard deviation or median and interquartile range. All statistical analyses were performed using the Stata software (version 17.0; StataCorp LP, College Station, TX, USA). All tests were two-tailed, and the threshold for significance was set at P < 0.05. The study was approved by the Institutional Review Board of the University of Tokyo (approval number: 3501-(5)), and the need for patient consent was waived because of the anonymous nature of the data.
RESULTS
During the study period, we identified 25,438 patients with ICD-10 diagnosis code G61.0, which is the code for GBS; among these patients, 1,595 (6.3%) had a specific diagnosis of MFS. Of these patients with MFS, 999 (62.6%) underwent immunotherapy. Specifically, 56.9% of patients with MFS received IVIG, 13.7% received intravenous methylprednisolone, and 1.8% received PE. The baseline characteristics of eligible patients stratified by immunotherapy administration are shown in Table 1. No differences in age, sex, body mass index, Japan Coma Scale score, Barthel index score, smoking status, or comorbidities were observed between the immunotherapy and non-immunotherapy groups. The proportion of unplanned admissions (86.9% vs. 74.3%) and admissions to academic hospitals (37.7% vs. 27.7%) among patients who received immunotherapy was higher than that among those who did not receive immunotherapy.
Table 1 Treatment and demographic characteristics of patients with Miller Fisher syndrome
|
Overall
(N = 1,595) |
Without immunotherapy
(N = 596) |
With
Immunotherapy
(N = 999) |
P |
| Immunotherapy, N (%) |
999 (62.6) |
|
|
|
| IVIG |
908 (56.9) |
— |
(90.9) |
|
| IVMP |
219 (13.7) |
— |
(21.9) |
|
| PE |
18 (1.8) |
— |
(2.9) |
|
| (Monotherapy) |
|
|
|
|
| IVIG alone |
762 (47.8) |
— |
(76.3) |
|
| IVMP alone |
77 (4.8) |
— |
(7.7) |
|
| PE alone |
11 (0.7) |
— |
(1.1) |
|
| (Dual therapy) |
|
|
|
|
| IVIG + IVMP |
132 (8.3) |
— |
(13.2) |
|
| IVIG + PE |
7 (0.4) |
— |
(0.7) |
|
| IVMP + PE |
3 (0.2) |
— |
(0.3) |
|
| (Triple therapy) |
|
|
|
|
| IVIG + IVMP + PE |
7 (0.4) |
— |
(0.7) |
|
| Age (years), N (%) |
|
|
|
0.343 |
| ≤19 |
52 (3.3) |
14 (2.4) |
38 (3.8) |
|
| 20–39 |
349 (21.9) |
141 (23.7) |
208 (20.8) |
|
| 40–59 |
539 (33.8) |
193 (32.4) |
346 (34.6) |
|
| 60–79 |
576 (36.1) |
217 (36.4) |
359 (35.9) |
|
| ≥80 |
79 (5.0) |
31 (5.2) |
28 (4.8) |
|
| Sex (female), N (%) |
672 (42.1) |
244 (40.9) |
428 (42.8) |
0.456 |
| Body mass index (kg/m2), N (%) |
|
|
|
0.893 |
| <18.5 |
148 (9.3) |
57 (9.6) |
91 (9.1) |
|
| 18.5–24.9 |
948 (59.4) |
347 (58.2) |
602 (60.2) |
|
| 25.0–29.9 |
327 (20.5) |
128 (21.5) |
199 (19.9) |
|
| ≥30.0 |
81 (5.1) |
32 (5.4) |
49 (4.9) |
|
| Missing |
91 (5.7) |
32 (5.4) |
59 (5.9) |
|
| Japan Coma Scale, N (%) |
|
|
|
0.334 |
| 0 (Alert) |
1,496 (93.8) |
565 (94.8) |
931 (93.2) |
|
| 1–3 (Awake without stimuli) |
91 (5.7) |
28 (4.7) |
63 (6.3) |
|
| 10–30 (Arousable with stimuli) |
6 (0.4) |
3 (0.5) |
3 (0.3) |
|
| 100–300 (Unarousable) |
2 (0.1) |
0 (0.0) |
2 (0.2) |
|
| Barthel index, N (%) |
|
|
|
0.089 |
| 95–100 (Independent) |
775 (48.6) |
304 (51.0) |
471 (47.2) |
|
| 65–90 |
252 (15.8) |
105 (17.6) |
147 (14.7) |
|
| 25–60 |
240 (15.1) |
81 (13.6) |
159 (15.9) |
|
| 0–20 |
107 (6.7) |
35 (5.9) |
72 (7.2) |
|
| Missing |
221 (13.9) |
71 (11.9) |
150 (15.0) |
|
| Smoking, N (%) |
634 (39.8) |
250 (42.0) |
384 (38.4) |
0.166 |
| Comorbidities at admission, N (%) |
|
|
|
|
| Diabetes mellitus |
175 (11.0) |
66 (11.1) |
109 (10.9) |
0.920 |
| Hypertension |
365 (22.9) |
136 (22.8) |
229 (22.9) |
0.962 |
| Chronic kidney disease |
13 (0.8) |
7 (1.2) |
6 (0.6) |
0.217 |
| Heart failure |
15 (0.9) |
9 (1.5) |
6 (0.6) |
0.069 |
| Unplanned admission, N (%) |
|
|
|
|
| Admission from home, % |
1,311 (82.2) |
443 (74.3) |
868 (86.9) |
<0.001 |
| Admission from other facilities, N (%) |
128 (8.0) |
75 (12.6) |
53 (5.3) |
<0.001 |
| Academic hospital, N (%) |
542 (34.0) |
165 (27.7) |
377 (37.7) |
<0.001 |
IVIG, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; PE, plasma exchange.
The distribution of immunotherapy use in hospitals is illustrated in Fig. 1. In total, 75 hospitals treated five or more cases of MFS, representing 5.3% of all hospitals that treated at least one patient with MFS. Among these, half of the hospitals (42.9%) administered immunotherapies to >80% of patients with MFS. In contrast, a small proportion of hospitals (4.0%) provided immunotherapy to <20% of patients with MFS.
Fig. 1
The histogram illustrates the distribution of immunotherapy use for Miller Fisher syndrome (MFS) across hospitals. The figure includes only hospitals with five or more admissions of MFA during the study periods. Among these, 42.9% of the hospitals administered immunotherapy to more than 80% of their patients with MFS.
An overview of the interventions and examinations performed during hospitalization and the initial diagnosis on admission is presented in Table 2. Patients who received immunotherapy were more likely to undergo tracheal intubation both during hospitalization (2.3% vs. 0.2%; P = 0.001) and within 5 days of admission (1.6% vs. 0.2%; P = 0.007), tracheotomy during hospitalization (1.4% vs. 0%; P = 0.004) and within 30 days of admission (1.3% vs. 0%; P = 0.005), as well as to receive antibiotics (17.0% vs. 7.9%; P < 0.001) and intensive care treatment during hospitalization (3.4% vs. 0.2%; P < 0.001) and within 5 days of admission (2.3% vs. 0.2%; P = 0.001) than those who did not receive immunotherapy. The proportion of patients who underwent lumbar puncture (75.5% vs. 62.8%; P < 0.001) and electromyography (71.4% vs. 55.0%; P < 0.001) was higher among those who received immunotherapy than among those who did not receive immunotherapy. Additionally, patients who received immunotherapy were more frequently diagnosed with GBS upon admission (2.6% vs. 0.8%; P = 0.014) than those who did not receive immunotherapy.
Table 2 Overview of intervention and examination among patients with Miller Fisher syndrome during hospitalization and initial diagnosis on admission, stratified by immunotherapy use
|
|
Overall
(N = 1,595) |
Without immunotherapy
(N = 596) |
With
(N = 999) |
P |
| Intervention, N (%) |
Tracheal intubation Mechanical ventilation, % |
|
|
|
|
| During hospitalization |
24 (1.5) |
1 (0.2) |
23 (2.3) |
0.001 |
| Within 5 days of admission |
17 (1.1) |
1 (0.2) |
16 (1.6) |
0.007 |
| Tracheotomy |
|
|
|
|
| During hospitalization |
14 (0.9) |
0 (0.0) |
14 (1.4) |
0.004 |
| Within 30 days of admission |
13 (0.8) |
0 (0.0) |
0 (1.3) |
0.005 |
| Antibiotic use |
217 (13.6) |
47 (7.9) |
170 (17.0) |
<0.001 |
| Intensive care unit stay |
|
|
|
|
| During hospitalization |
35 (2.2) |
1 (0.2) |
34 (3.4) |
<0.001 |
| Within 5 days of admission |
24 (1.5) |
1 (0.2) |
23 (2.3) |
0.001 |
| Examination, N (%) |
Lumbar puncture |
1,128 (70.7) |
374 (62.8) |
754 (75.5) |
<0.001 |
| Electromyogram |
1,041 (65.3) |
328 (55.0) |
713 (71.4) |
<0.001 |
| Initial diagnosis on admission, N (%) |
Other than Miller Fisher syndrome |
338 (21.2) |
137 (23.0) |
201 (20.1) |
0.175 |
| Guillain–Barré syndrome |
31 (1.9) |
5 (0.8) |
26 (2.6) |
0.014 |
| Bickerstaff syndrome |
1 (0.1) |
0 (0.0) |
1 (0.1) |
0.440 |
| Ischemic stroke |
34 (2.1) |
15 (2.5) |
19 (1.9) |
0.411 |
| Meningitis/encephalitis |
14 (0.9) |
5 (0.8) |
9 (0.9) |
0.898 |
| Myasthenia gravis |
9 (0.6) |
1 (0.2) |
8 (0.8) |
0.103 |
| Peripheral/vestibular vertigo without specific diagnosis |
22 (1.4) |
11 (1.9) |
11 (1.1) |
0.217 |
| Eye movement disorders without specific diagnosis |
48 (3.0) |
23 (3.9) |
25(2.5) |
0.125 |
| Ataxia without specific diagnosis |
17 (1.1) |
8 (1.3) |
9 (0.9) |
0.406 |
The outcomes of hospitalization are shown in Table 3. The length of hospital stay was significantly longer among patients who received immunotherapy (median [interquartile range], 18 [13–29] days) than among those who did not receive immunotherapy (median [interquartile range], 12 [8–21] days; P < 0.001). The total hospitalization cost for patients receiving immunotherapy was significantly higher than that for those not receiving immunotherapy (median, 1,660,200 JPY vs. 549,375 JPY; P < 0.001). No significant differences were observed in in-hospital mortality, Japan Coma Scale scores at discharge, or Barthel index scores at discharge between the groups. Most patients were discharged with fully alert consciousness, described as the Japan Coma Scale score of 0 (97.5% and 97.3%), and independent ADL described as the Barthel index score of 95–100 (75.3% and75.2), whereas a small portion of patients experienced a deteriorated Barthel index score during hospitalization (4.0% and 4.2%) in both groups with and without immunotherapies.
Table 3 Outcomes of patients with Miller Fisher syndrome during hospitalization, stratified by immunotherapy use
|
Overall
(N = 1,595) |
Without immunotherapy
(N = 596) |
With Immunotherapy
(N = 999) |
P |
| Length of hospital stay, days (interquartile range) |
16 (11–26) |
12 (8–21) |
18 (13–29) |
<0.001 |
| In-hospital death, median, N (%) |
8 (0.5) |
1 (0.2) |
7 (0.7) |
0.145 |
| Total medical cost, median, JPY |
1,457,570 |
549,375 |
1,660,200 |
<0.001 |
| Japan Coma Scale at discharge, N (%) |
|
|
|
0.981 |
| 0 (Alert) |
1,554 (97.4) |
580 (97.3) |
974 (97.5) |
|
| 1–3 (Awake without stimuli) |
33 (2.1) |
15 (2.5) |
18 (1.8) |
|
| 10–30 (Arousable with stimuli) |
(0.0) |
(0.0) |
(0.0) |
|
| 100–300 (Unarousable) |
(0.0) |
(0.0) |
(0.0) |
|
| Missing |
8 (0.5) |
1 (0.2) |
7 (0.7) |
|
| Barthel index at discharge, N (%) |
|
|
|
0.219 |
| 95–100 (Independent) |
1,200 (75.2) |
448 (75.2) |
752 (75.3) |
|
| 65–90 |
172 (10.8) |
64 (10.7) |
108 (10.8) |
|
| 25–60 |
73 (4.6) |
26 (4.4) |
47 (4.7) |
|
| 0–20 |
34 (2.1) |
12 (2.0) |
22 (2.2) |
|
| Missing |
116 (7.3) |
46 (7.7) |
70 (7.0) |
|
| Comparison of Barthel index between admission and discharge, N (%) |
|
|
|
0.164 |
| 95–100 on both admission and discharge |
712 (44.6) |
280 (47.0) |
432 (43.2) |
|
| Improved by 5 and more |
485 (30.4) |
174 (29.2) |
31.1 (31.1) |
|
| Unchanged |
55 (3.5) |
26 (4.4) |
29 (2.9) |
|
| Deteriorated by 5 and more |
65 (4.1) |
25 (4.2) |
40 (4.0) |
|
| Missing |
278 (17.4) |
91 (15.3) |
187 (18.7) |
|
DISCUSSION
In this study, 62.6% of the patients with MFS in Japan received immunotherapy. Most patients had independent ADL at discharge, regardless of immunotherapy use, suggesting an overall favorable prognosis. Although the background characteristics of the patients were not adjusted, those who received immunotherapy had longer hospital stays and higher hospitalization costs than those who did not receive immunotherapy. No significant differences in mortality, level of consciousness, or ADL at discharge were found between the groups.
Our study provides an overview of the treatments administered to patients with MFS in Japan, revealing that more than half of them received immunotherapies, predominantly IVIG, and less commonly, PE. Previous studies have suggested that MFS follows a generally favorable natural course. A retrospective cohort study of 92 patients with MFS (28 receiving IVIG, 23 receiving PE, and 41 receiving no immunotherapy) revealed no differences in the improvement of external ophthalmoplegia and ataxia among the treatment groups3), supporting this favorable outcome3),4). Another study that tracked the natural history of 50 consecutive patients with MFS revealed that ataxia resolved within an average of 1 month, external ophthalmoplegia resolved within 3 months, and only two patients exhibited diplopia at 6 months of onset2). Our findings highlight the prevalent use of immunotherapies in MFS management despite the potential for spontaneous recovery in most cases.
We speculate that the clinical presentation of each patient may have influenced clinicians’ decisions regarding immunotherapy. In our study, the baseline characteristics, including sex, age, weight, ADL, and consciousness at admission, were comparable between the immunotherapy and non-immunotherapy groups. However, a high proportion of patients who received immunotherapy required tracheal intubation, tracheotomy, antibiotics, and intensive care unit admission, suggesting a severe clinical course in this group. The Japanese clinical guidelines for MFS recommend considering immunotherapy in patients with rapid-onset complete external ophthalmoplegia4). However, the database does not contain specific clinical details. Factors closely related to the severity of MFS, which were not captured in our dataset, may have influenced the decision to initiate immunotherapy.
In our cohort, the overall prognosis of MFS was generally favorable, as evidenced by independent ADL in 75% of patients, with only 4.1% experiencing deteriorated ADL during hospitalization. ADL at discharge was somewhat less favorable than that reported in other studies, where most patients exhibited complete recovery3),4). This discrepancy may be partly attributed to the shorter observation period in our study; the median hospital stay was 16 days, and a longer observation period may have captured more patients with spontaneous recovery. Additionally, no significant differences in in-hospital death, ADL at discharge, level of consciousness at discharge, and changes in ADL were observed between the immunotherapy and non-immunotherapy groups. Because immunotherapy was more frequently administered to patients with more severe conditions, these comparable outcomes suggest that immunotherapy may have contributed to clinical recovery despite greater initial severity. However, these findings should be interpreted with caution because we were unable to adjust for potential confounders.
Most hospitals administer immunotherapy to >80% of patients with MFS, suggesting its widespread use as a de facto standard in many medical facilities in Japan. Immunotherapies such as IVIG and PE rely heavily on substantial human blood resources and incur high medical costs. In addition, the increasing demand for these treatments has led to a shortage of immunoglobulin products13). Given these considerations, immunotherapy may be more appropriate for patients who present with severe neurological symptoms and those at high risk for deterioration, including progression to GBS or Bickerstaff’s brainstem encephalitis.
This study has several limitations. Because this study used a database that was not originally constructed for research purposes, the recorded MFS diagnoses may not have been accurate. Some patients in our cohort exhibited atypical clinical manifestations of MFS, such as impaired consciousness and respiratory failure, and required mechanical ventilation or tracheotomy. This finding suggests that differentiating MFS from GBS or Bickerstaff’s brainstem encephalitis can be challenging in some patients, potentially influencing the decision to administer immunotherapy. Additionally, the lack of key clinical variables in the database, such as detailed neurological findings, respiratory function, and the department to which the patient was admitted, hindered the necessary background adjustments to compare the outcomes. This limitation highlights the need for future prospective studies incorporating detailed measures of disease severity to accurately evaluate the effectiveness of immunotherapy for MFS. Despite these limitations, we collected large-scale nationwide data on MFS, which is a relatively rare condition, highlighting the widespread use of immunotherapies for MFS in Japan despite its generally favorable natural course.
CONCLUSION
Although MFS is generally associated with a self-limiting clinical course and favorable neurological outcomes, immunotherapies such as IVIG, PE, and intravenous methylprednisolone are widely used for this condition in Japan.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest in relation the work presented in the manuscript.
FUNDING SOURCES
This work was supported by grants from the Ministry of Health, Labor and Welfare, Japan (23AA2003 and 24AA2006).
AUTHOR CONTRIBUTIONS
Satoshi Kodama contributed to study design, data collection, data analysis, data interpretation and writing the manuscript. Mitsuhiro Kainaga, Shotaro Aso, Taisuke Jo, Yohei Hashimoto, Hiroki Matsui, Hideo Yasunaga, and Kiyohide Fushimi contributed to study design, data collection, data analysis, data interpretation and critical revision of the manuscript. Yuichiro Shirota, Masashi Hamada and Tatsushi Toda contributed to data interpretation and critical revision of the manuscript. All authors reviewed and approved the final draft of the manuscript.
DISCLAIMER
Hideo Yasunaga is one of the Editorial Board members of Annals of Clinical Epidemiology (ACE). This author was not involved in the peer-review or decision-making process for this paper.
References
- 1. Mori M, Kuwabara S. Fisher syndrome. Curr Treat Options Neurol 2011;13:71–78.
- 2. Mori M, Kuwabara S, Fukutake T, Yuki N, Hattori T. Clinical features and prognosis of Miller Fisher syndrome. Neurology 2001;56:1104–1106.
- 3. Mori M, Kuwabara S, Fukutake T, Hattori T. Intravenous immunoglobulin therapy for Miller Fisher syndrome. Neurology 2007;68:1144–1146.
- 4. Japanese Society of Neurology. Practical guideline for Guillain barre syndrome and Fisher syndrome 2013. Tokyo: Nankodo Co, Ltd; 2013.
- 5. Sekiguchi Y, Mori M, Misawa S, Sawai S, Yuki N, Beppu M, et al. How often and when Fisher syndrome is overlapped by Guillain–Barré syndrome or Bickerstaff brainstem encephalitis? Eur J Neurol 2016;23:1058–1063.
- 6. Verboon C, van Berghem H, van Doorn PA, Ruts L, Jacobs BC. Prediction of disease progression in Miller Fisher and overlap syndromes. J Peripher Nerv Syst 2017;22:446–450.
- 7. Gold R, Stangel M, Dalakas MC. Drug Insight: the use of intravenous immunoglobulin in neurology—therapeutic considerations and practical issues. Nat Clin Pract Neurol 2007;3:36–44.
- 8. Yasunaga H, Matsui H, Horiguchi H, Fushimi K, Matsuda S. Clinical epidemiology and health services research using the diagnosis procedure combination database in Japan. Asian Pac J Dis Manag 2015;7:19–24 (in Japanese).
- 9. Yamana H, Matsui H, Sasabuchi Y, Fushimi K, Yasunaga H. Categorized diagnoses and procedure records in an administrative database improved mortality prediction. J Clin Epidemiol 2015;68:1028–1035.
- 10. Yamana H, Moriwaki M, Horiguchi H, Kodan M, Fushimi K, Yasunaga H. Validity of diagnoses, procedures, and laboratory data in Japanese administrative data. J Epidemiol 2017;27:476–482.
- 11. Nakajima M, Okada Y, Sonoo T, Goto T. Development and validation of a novel method for converting the Japan Coma Scale to Glasgow Coma Scale. J Epidemiol 2023;33:531–535.
- 12. Shigematsu K, Nakano H, Watanabe Y. The eye response test alone is sufficient to predict stroke outcome—reintroduction of Japan Coma Scale: a cohort study. BMJ Open 2013;3:e002736.
- 13. Roth K, Darwish C, Keller MD, Hammer B, Ahmed-Winston S, Escalante E, et al. Implementation of a tier system for IVIG indications to address IVIG shortage at a tertiary care pediatric medical center. Pediatr Blood Cancer 2024;71:e30871.
