2019 JAPAN Critical Limb Ischemia Database (JCLIMB) Annual Report

Since 2013, the Japanese Society for Vascular Surgery has started the project of nationwide registration and tracking database for patients with critical limb ischemia (CLI) who are treated by vascular surgeons. The objective of this project is to elucidate the current status of the medical practice for CLI patients to contribute to the improvement of the quality of medical care. This database, called JAPAN Critical Limb Ischemia Database (JCLIMB), is created on the National Clinical Database (NCD) and collects data of patients’ background, therapeutic measures, early results, and long-term prognosis as long as 5 years after the initial treatment. The limbs managed conservatively are also registered in JCLIMB, together with those treated with surgery and/or endovascular treatment (EVT). In 2019, 1070 CLI limbs (male: 725 limbs, 68%) were registered by 83 facilities. Arteriosclerosis obliterans (ASO) accounted for 98% of the pathogenesis of these limbs. In this manuscript, the background data and the early prognosis of the registered limbs are reported. Although the registration format for the simultaneous surgery of bilateral limbs in NCD was changed to one patient and two limbs, JCLIMB still counted two patients and two limbs to eliminate discrepancy with the past annual reports. (This is a translation of Jpn J Vasc Surg 2022; 31: 157–185.)


Introduction
Recently, an increasing number of patients with critical limb ischemia (CLI) have undergone medical care at clinical practice sites. Improving the treatment outcome for these patients is an important and urgent issue. Since 2013, the Japanese Society for Vascular Surgery (JSVS) has initiated the project of a nationwide CLI registration and tracking database to obtain CLI epidemiological data that can be shared among the medical staff. The background of CLI limbs, treatment contents, early outcome, and long-term outcome until 5 years after surgery, including non-surgical limbs, are registered in this database. The database was named JAPAN Critical Limb Ischemia Database (JCLIMB) and established on the National Clinical Database (NCD). The primary objective of the JCLIMB project is to elucidate the current status of CLI treatment performed by vascular surgeons in Japan and inform physicians at practice sites of those, thus improving the quality of medical care. The initial registration data, and their tracking data 1 month after registration in 2013-2018, have already been published. [1][2][3][4][5][6] This article reports the basic data registered in 2019.

JCLIMB
Registration details, including the definition of CLI, have already been described in the 2013 annual report. 1) The CLI to be registered was defined according to TASC II 7) : chronic ischemic rest pain, ulcer, or gangrene attributable to objectively proven arterial occlusive disease. The CLI diagnosis should be confirmed by ankle pressure (AP) below 50 mmHg or toe pressure (TP) below 30 mmHg in limbs with rest pain and by AP below 70 mmHg or TP below 50 mmHg in limbs with ulcer or gangrene.
The same limb can be registered in JCLIMB only once within a 5-year tracking period. When the registered limb is treated at different times or at different institutions, such data should be added only to the tracking items of each limb in JCLIMB, avoiding registration overlap. However, details of the procedure are registered each time in the NCD apart from the registration in JCLIMB. On the other hand, the patient with bilateral CLI can be registered twice for each limb. Based on the NCD regulations, fixing of JCLIMB is done as follows: Initial registration data: Early April in the following year, tracking data early after treatment (1 month)/6 months after treatment: end of December in the following year, tracking data 1 year after treatment: end of December after 2 years.
Tracking data 2, 3, 4, and 5 years after treatment were registered until the end of December after 3, 4, 5, and 6 years, respectively.
As a general rule, the timing of tracking data registration is accepted within a ±2-month range until 12 months after treatment and within a ±3-month range thereafter. Although the day for tracking data fixing is specified, it is made flexible because in some limbs, follow-up data might be revealed later.
It is very difficult to require facilities participating in NCD to register CLI data because a significant number of registration items in JCLIMB would put too much burden on them. Thus, facilities wishing to participate were recruited. In total, 83 facilities, which registered CLI limbs in 2019 at the time of compiling in December 2021, are listed in the appendix.
Because JCLIMB is considered a registry study on NCD, patient consent to participate in the study and the ethical review of the study at the time of participation in NCD were adopted.

Comments on the Aggregated Data in 2019
The initial registration data in 2019 were fixed in early April 2020, and the tracking data early after treatment (1 month) were fixed in April 2021. In December 2021, 1070 limbs, belonging to 725 males (68%) and 345 females (32%), were registered in 83 facilities. All data and extracted data on arteriosclerosis obliterans (ASO) were collected according to the registered items. Because ASO accounted for 98% of all limbs, the overall and ASO data exhibited similar tendencies. In the comments, ASO data were presented in parentheses. In addition, because the WIfI classification of the Society for Vascular Surgery (SVS) was reported in 2014 (Tables 1-1-1-1-1-3), 8) JCLIMB made several changes and additions to the registered items, making the WIfI classification possible since 2015 (Tables 1-2-1-1-2-3). The total figure was not always consistent, mostly due to missing values, and an explanation for each inconsistency was added. Although the registration format for bilateral simultaneous surgery was changed from two limbs in two patients to two limbs in one patient in the NCD in July 2019, the data was calculated on a limb basis as before in the 2019 JCLIMB annual report to eliminate discrepancy with the past reports.

(1) Pretreatment patients background
The pretreatment patients background is presented in Tables 2-1-2-6. Good blood pressure control was defined as blood pressure below 140/90 mmHg in the absence of diabetes and renal failure or blood pressure below 130/80 mmHg in the presence of these diseases. Good diabetes control was defined as hemoglobin A1c below 7.0% (National Glycohemoglobin Standardization Program [NGSP] value). Good dyslipidemia control was defined as low-density lipoprotein below 100 and 80 mg/dL in the absence and presence of other arteriosclerotic diseases, respectively. The presence of heart failure was judged clinically. The patient was regarded as having heart failure based on a past history of admission due to heart failure, clinical symptoms of heart failure, diagnosis of heart failure confirmed via echocardiography, or reduced cardiac function as revealed by echocardiography even with no clinical heart failure symptoms. Renal dysfunction was graded according to the new chronic kidney disease severity classification of the Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease 2012 9) : Renal dysfunction was absent when the estimated glomerular filtration rate (eGFR) (mL/min/1.73 m 2 ) was 60 or higher, and it was graded as G3a, G3b, G4, and G5 when the eGFR was 45-59, 30-44, 15-29, and below 15, respectively. eGFR below 15 in hemodialysis patients was graded as G5D.
The problems and considerations on these spreadsheets are described below. In Table 2-4, the total number of malignant neoplasm sites is larger than that of malignant neoplasms. This is because multiple selections of malignant neoplasm sites were possible. The blood flow data (ankle brachial index, toe brachial index [TBI], and skin perfusion pressure [SPP]) of contralateral limbs were omitted in this report as there were many missing values, though they had been displayed until 2018.

(2) Conditions of limb ischemia
Limb ischemia pretreatment conditions are presented in Tables 3-1 to 3-6. Regarding the walking function (Taylor classification), 10) patients who could walk outdoors or indoors independently, including with a cane, were regarded as ambulatory, whereas those unable to walk but able to The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team stand on their own legs during transfer from the bed to a wheel chair were designated as ambulatory/homebound.
Regarding the state of local tissue defect (Texas University Classification), 11) the most severe lesion, the main treatment target, was evaluated. SPP was measured on the foot (base of the toe, dorsum of the foot, or sole), and a lower value was used. To perform the WIfI classification, the ulcer and gangrene sites were separately registered. Although SPP is widely used as an objective index for evaluating ischemia in Japan, the ischemic grading criteria using SPP is not shown in the WIfI classification, in which TP is given top priority. Therefore, in JCLIMB, the SPP value was converted to TP using the conversion equation SPP= 0.6853× TP +14.48 from the correlation data of SPP and TP reported in Japan 12) and applied for WIfI ischemic grading (Table 1-2-2). Because the ischemic grade of the WIfI classification based on the SPP value was newly defined in the Japanese PAD guideline revised in 2022, the classification method will be changed in the annual report after 2020.
The lesion was considered infected when it showed two or more of the following symptoms: local swelling or induration, erythema >0.5 cm around the ulcer, local tenderness or pain, local warmth, and purulent discharge (thick, opaque to white, or sanguineous secretion). In addition, local infections involving only the skin and the subcutaneous tissue, and those involving structures deeper than the skin and subcutaneous tissues, were separately registered. Local infections involving only the skin and the subcutaneous tissue were differentiated based on the size of the erythema around the ulcer, ≦2 or >2 cm. Systemic inflammatory response syndrome, indicating systemic infection, was manifested by two or more of the following symptoms: temperature >38°C or <36°C, heart rate >90 beats/min, respiratory rate >20 breaths/min or PaCO 2 <32 mmHg, white blood cell count >12,000 or <4,000 cu/mm, or 10% immature (band) forms. The arteries in the ankle joint region were classified as foot arteries.
The problems and considerations on these spreadsheets are described below. In Table 3-1, the total number of ambulatory function differed from the total number of the main sites of ulcer/gangrene to be treated. This is because there were missing values in the main sites of ulcer/gangrene to be treated. In Table 3-3, the total number of limbs in the TASC II classification differed from the number in each column of the site of occlusion. In the aortoiliac lesion, a decreased number of that in the TASC II classification may have been due to input omission. In the femoropopliteal lesion, an increased number of that in TASC II may have been due to the crural lesions. In Table 3-6, there was some dissociation between the R and Wound grades. This may be due to the R grade s obscure definition. For example, extensive gangrene involving the forefoot is classified in R5 and W3, whereas a shallow ulcer without exposure of the distal leg bone is classified in R6 and W1. In Table 3-6, 82 limbs (81 limbs) were registered as ischemic grade 0 in the WIfI classification. By definition, a limb with ischemic grade 0 has an ABI 0.8 or higher or AP higher than 100 mmHg, or if arterial calcification precludes reliable ABI or AP measurements, TP of 60 mmHg or higher or TcPO 2 60 mmHg or higher (SPP 55 mmHg or higher in JCLIMB) (Table 1-1-2). There should be no limb with ischemic grade 0 because the CLI registered in JCLIMB is defined according to TASC II. The limbs might be clinically judged to be CLI irrespective of the objective ischemic index, although details are unknown. Table  3-6 demonstrates that there were three limbs (3 limbs) in which infection was confirmed in R4 limbs, despite the absence of a local wound by definition of R4. This may occur because tissue loss is not always requisite for fI grade. In Table 3-6, the numbers of wound, ischemia, foot infection, and stage are different. This is because there were missing values in items required for grading.

(3) Treatment
Tables 4-1 to 4-6 present the CLI treatment data. Revascularizations of the affected limbs were performed in 94% (95%) of the registered limbs, and primary major amputations were performed in 2.3% (2.4%) of the registered limbs. Among the surgical reconstruction procedures, distal bypass accounted for 54% (54%). Endovascular treatment (EVT), including EVT alone and hybrid treatment with surgical reconstruction, accounted for 54% (55%) of the total revascularization procedures. The EVT applied to the crural or foot artery accounted for 37% (36%) of the total EVT.
The problems and considerations on these spreadsheets are described below. In Table 4-1, the sum of the number of cells in treatment is larger than that of the number of registered limbs, 1070 (1049), because more than one treatment method can be selected. In Table 4-1, the discrepancy in the number of major amputation to the number of detail of amputation was caused by unused. In the column of vein usage of Table 4-3, how the autologous veins were used was described when they were selected as vascular conduits. The sum of the number in the column of vein usage, in-situ, non-reversed, reversed, spliced, and patch, is larger than the sum of the number in the column of vein in vascular prosthesis. It could be because of selecting multiple vein usage for arterial reconstruction in a limb. The sum of the number in the column of vein in vascular prosthesis is identical to the sum of the number in the column of vein quality. Vascular prosthesis (−) included an endarterectomy without a patch angioplasty. In Table 4-4, the sum of the number of proximal anastomosis is not equal to the sum of the number of distal anastomosis. This was because multiple arteries could be selected in each anastomosis. The total number of distal anastomosis sites of the foot artery is larger than that of distal anastomosis foot. This was because multiple sites were selected in dual bypass. Table 4-6 summarizes the vascular grafts used for the infrainguinal arterial reconstruction. For example, the total number of vascular graft in the column of femoralproximal popliteal artery bypass was 50 (49), which was higher than 47 (46), the number of actual applications in Table 4-2. This was because multiple graft materials could be selected when multiple procedures, such as a sequential bypass procedure and TEA, can be performed simultaneously for arterial reconstruction in the lower limb.

(4) Outcomes early (1 month) after treatment
Tables 5-1 to 5-8 present the outcomes early (1 month) after treatment. At the time of summary count at the end of April 2021, follow-up data 1 month after treatment were obtained in 886 limbs (83%), including 866 limbs (83%) with ASO. Data were collected according to the severity of the local limb conditions (Rutherford classification) and treatment measures (EVT alone or surgical reconstruction with/without EVT). The mortality rate was 3.6% (3.7%) in the whole series and 3.8% (3.9%) and 2.9% (3.0%) treated with EVT alone and with surgical reconstruction with/without EVT, respectively. The most common cause of death was cardiac disease, which accounted for 34% (34%) of all deaths. Postoperative complications were cardiac disease in 3.0% (3.0%), cerebrovascular disease in 1.8% (1.7%), pneumonia in 2.4% (2.4%), and wound complication in 4.5% (4.5%). Complications at the puncture site were noted in 1.0% (1.0%) of the limbs treated with EVT alone.
The problems, comments, and considerations on these spreadsheets are described below. The number of bypass graft/EVT condition, clinical limb symptoms, ischemic wound, and ambulatory function at discharge did not match ( Table 5-5). The total number of ambulatory function at discharge was 886 (866), which was equal to the number of life prognoses ( Table 5-1), indicating no unused. The number of bypass graft/EVT condition was not equal to the number of ambulatory function at discharge because the objective of bypass graft/EVT condition was to achieve survival with arterial reconstruction of the limbs and because more than one condition could be selected. The numbers of clinical symptoms of limb and ischemic wound were not identical. They must be identical because their objective was to achieve survival without major amputations. This is speculated to be due to the presence of unused in dead cases before registration. The discrepancy in the total number of life prognosis, clinical limb symptom, and amputation is due to the difference of condition for data aggregation. In Table  5-3, the presence of the puncture site complication in nonreconstruction group seems to be odd. The registration of complication at the puncture site was required in limbs where PTA/STENT was selected in the revascularization method. However, in JCLIMB, multiple treatment methods other than revascularization were selected, which caused the odd results. It is presumed to be due to input error or EVT failure.
The number of limbs of survivors with EVT was 405 (399 limbs) (Table 5-1), which was 9 (9) limbs less than the sum of the number in the column of minor reintervention or major reintervention in the row of limbs with EVT, 414 limbs (408 limbs) ( Table 5-6). The number of limbs of survivors with surgical reconstruction was 401 (389 limbs) ( Table 5-1), which was 8 (8) limbs less than the sum of the number in the column of minor reintervention or major reintervention in the row of limbs with surgical reconstruction, 409 limbs (397 limbs) ( Table 5-6). This is speculated to be due to death after reintervention. In Table  5-6, the objective for input of revision for those excluding good bypass graft/EVT condition is limb registered in stenosis, occlusion, deterioration, anastomosis disruption (aneurysm), infection, and others of bypass graft/EVT condition. The number of bypass graft/EVT condition of surgical reconstruction and total in Table 5-5 does not match the number of revision for those excluding good bypass graft/EVT condition in Table 5-6. This is because multiple items can be selected in bypass graft/EVT condition, and both infection and anastomosis disruption (aneurysm) were selected in a case. The total number of the contralateral limb occlusive lesions in Table 5-7 is 18 limbs (18 limbs) less than life prognosis in Table 5-1 due to missing values. The sum of the number of treatment for contralateral limb is less than that of the contralateral limb occlusive lesions as the objectives of treatment for contralateral limb excluded the limbs with no occlusive lesions in the contralateral limb. Because multiple registrations were possible, the sum of the number of treatment for contralateral limb was more than that of the limbs with occlusive lesions in the contralateral limb. When a patient died within 1 month, the information of newly diagnosed malignant neoplasm at death was registered in Table 5-8.
In addition to the above, there were some parts where the total number does not match in Tables 5-1 to 5-8. It might be because several items had multiple choices or missing values.

Conclusions
Vascular surgeons contribution to the participating facilities is the sufficient amount of detailed data during busy clinical practice, which has gradually elucidated the current status of CLI treatment in Japan. Data on CLI in 2018 were elucidated, after the annual data in 2013-2017. The JCLIMB Committee is planning to continue publishing an annual report in the future. In 2017, the new concept, chronic limb threatening ischemia, was proposed instead of CLI, 14) and a new clinical guideline, the Global Vascular Guideline, was published instead of TASC in 2019. 15) The full name of JCLIMB has been changed to Japan Chronic Limb-Threatening Ischemia Database, and the data format has been revised to register the survey items according to the Global Vascular Guideline, which can be used in 2021.
The JCLIMB Committee expects that these study results will be fed back to clinical situations to help develop medical care for CLI. The paper regarding 30 days prediction model using the data of JCLIMB has been published, 16) and the paper regarding 2 years prediction model was submitted. Facilities can participate in JCLIMB at any time by contacting the JSVS secretariat for details.
• CLI was used in this paper because the objectives registered in 2019 were based on CLI defined by TASC II.

Disclosure Statement
The authors have no conflict of interest.

Additional Remarks
This report was authorized by the institutional review board of Saiseikai Yahata General Hospital. (Authorization No.185)

Additional Note
The original Annual Report was published in Japanese Journal of Vascular Surgery Vol. 31 (2022) No. 3; however, errors in numerical data were detected after the publication. The errata were published in Vol. 31 (2022) No. 5 of the same journal. This translation reflects that correction.  The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team Involving only the skin and the subcutaneous tissue (Erythema around the ulcer; 0.5-2 cm) Involving only the skin and the subcutaneous tissue (Erythema around the ulcer; >2 cm), or involving structures deeper than skin and subcutaneous tissues (e.g., abscess, osteomyelitis, septic arthritis, fasciitis)     The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team   The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team  cm around the ulcer, Local tenderness or pain, Local warmth, Purulent discharge (thick opaque to white, or sanguineous secretion) Local infection are skin and subcutaneous tissue was classified by the spreading of erythema (≦2 cm or >2 cm around the ulcer/gangrene) Local infection involving structures deeper than skin and subcutaneous tissues (e.g., abscess, osteomyelitis, septic arthritis, fasciitis) The signs of SIRS are manifested by two or more of the following: Temperature >38°C or <36°C, Heart rate >90 beats/min, Respiratory rate >20 breaths/min or PaCO   The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team       The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team      The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team    The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team    The Japanese Society for Vascular Surgery JCLIMB Committee, NCD JCLIMB Analytical Team