Incidence rates for thromboembolic, bleeding and hepatic outcomes in patients undergoing hip or knee replacement surgery

Authors


Stephan Lanes, United BioSource Corporation, Center for Epidemiology and Database Analytics, 430 Bedford St, Suite 300, Lexington, MA 02420, USA.
Tel.: +1 203 405 1552; fax: +1 203 405 1552.
E-mail: stephan.lanes@unitedbiosource.com

Abstract

Summary. Background: Data on clinical outcomes of patients in the general population undergoing knee replacement or hip replacement surgery are sparse. Objectives: To conduct an observational study using insurance claims data to assess the incidence of selected clinical events following knee replacement or hip replacement surgery in the USA. Patients/Methods: A total of 97 469 knee replacement patients and a total of 45 203 hip replacement patients were included during the period 2004–2008; the median age was 64 years, and 63% of knee replacement patients and 55% of hip replacement patients were women. Results: During a median follow-up of 70–71 days, the incidence rates in knee replacement patients and hip replacement patients were, respectively: ischemic stroke, 15 and 19 per 1000 person-years; acute coronary syndrome (ACS), 15 and 18 per 1000 person-years; bleeding events, 46 and 47 per 1000 person-years; venous thromboembolism (VTE), 64 and 45 per 1000 person-years; and hepatic events, one and one per 1000 person-years. Approximately 45% of knee replacement and hip replacement patients had no claims for outpatient anticoagulant therapy within 1 week after discharge from hospital. Conclusions: Ischemic events such as stroke, ACS and VTE are important adverse events following knee replacement and hip replacement surgery. The results reported here can help in making challenging decisions regarding the clinical management of risks attributable to bleeding events and clotting events.

Introduction

Knee and hip replacement operations are rapidly increasing in frequency. According to the US National Hospital Discharge survey conducted between 1990 and 2004, there were 0.8 million primary total knee arthroplasties in1990–1994, 1.2 million in 1995–1999, and 1.8 million in 2000–2004 [1]. In 2004, approximately 400 000 primary knee replacements and 200 000 primary hip replacements were performed in the USA, representing increases of 53% and 37% as compared with the year 2000 [2]. It has been estimated that, in 2010, approximately 720 000 knee replacements and 300 000 hip replacements would be performed in the USA [3]. Significant numbers of hip and knee arthroplasties have also been observed in other developed countries [3].

Both hip and knee arthroplasties are associated with an immediate and, in the case of hip arthroplasty, prolonged hypercoagulable state with an increased risk of venous thromboembolism (VTE), comprising deep vein thrombosis (DVT) and pulmonary embolism (PE). Data from clinical trials showed that, 7–14 days after hip or knee arthroplasty, 40–60% of patients developed DVT and 1–10% developed PE when receiving either no thromboprophylaxis or placebo [4]. These findings underscore the importance of prophylactic measures for the prevention of VTE in this patient population.

Anticoagulants as a means to prevent VTE have been shown to be efficacious in clinical trials. Accordingly, the American College of Chest Physicians (ACCP) recommends thromboprophylaxis with anticoagulants for a minimum of 10 days for patients undergoing hip or knee arthroplasty,, and for up to 35 days for those undergoing hip arthroplasty [4]. The American Academy of Orthopaedic Surgeons (AAOS) and the Surgical Care Improvement Project also recommend prophylactic measures in these patients [5,6]. Little is known about how well these guidelines are followed, or the frequencies of different patient outcomes in real-world settings [7]. We analyzed a large US medical/pharmacy claims database to assess the postsurgery incidence rates for important clinical events in patients undergoing hip or knee replacement surgery.

Materials and methods

We conducted a retrospective database study using the MarketScan® Commercial Claims and Encounters Database, from Thomson Reuters (New York, NY, USA). The MarketScan database is a Health Insurance Portability and Accountability Act of 1996-compliant, fully integrated, de-identified patient-level medical insurance claims database containing enrollment dates as well as data about demographic and benefit design information and claims for inpatient visits, outpatient visits, pharmacy, laboratory tests, and health risk assessment services from commercial and Medicare supplemental insurance for more than 36 million patients representative of the geographic distribution of the US population.

We created two study cohorts of patients undergoing surgery for knee replacement and hip replacement from 1 July 2004 to 30 June 2008. Patients had at least one inpatient claim with an International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) procedure code, Healthcare Common Procedure Coding System code or Common Procedural Terminology (CPT, version 4) code representing hip replacement surgery (ICD-9-CM 81.51, 81.52, and 81.53; CPT 27125, 27130, 27132, 27134, 27137, and 27138) or knee replacement surgery (ICD-9-CM 81.54 and 81.55; CPT 27437–27447 and 27486–27487). All patients had to be at least 18 years of age and have had at least 180 days of continuous enrollment in the database before their index date. The index date was defined as the date of the surgical procedure or the date of first anticoagulant therapy within 7 days before the procedure, whichever came first. The baseline period was defined as the 6-month period preceding the index date. Patients were excluded if, during the baseline period, they had any primary or secondary inpatient or outpatient diagnosis codes indicating bleeding or hepatobiliary disease, or were treated with any of the following classes of medication: anticoagulants, fibrinolytics, antiplatelet agents, or antifibrinolytics. In addition, for the analysis of each endpoint, we excluded prevalent cases by excluding patients with any diagnosis codes for the endpoint of interest during the baseline period.

Baseline characteristics for patients in each study cohort were identified from medical and pharmacy claims during the baseline period. We classified patients with respect to age, gender, region of the USA, and year of index date. In addition, patients were classified according to comorbidities that may be related to risk of one of the study endpoints. Comorbidity and risk factor diagnoses and procedures were identified from either inpatient or outpatient claims, and could be from primary and non-primary diagnosis fields.

Subjects were followed from their index day until the earliest occurrence of the following: (i) 60 days after the end of anticoagulant treatment; (ii) end of enrollment; (iii) calendar end of the study (30 June 2008); (iv) 365 days from the index day (for those with continuous anticoagulant therapy treatment); or (v) the first occurrence of each endpoint. In addition, patients were censored from the analysis if they were not discharged within 10 days after the surgical procedure.

Study endpoints included: hemorrhagic stroke (ICD-9-CM 430.x, 431.x), ischemic stroke (ICD-9-CM 433.x, 434.x, and 436.x, excluding 433.00 and 434.00), acute coronary syndrome (ACS) (ICD-9-CM 410.x and 411.x), VTE (ICD-9-CM 451.x, 453.x, and 415.1), PE (ICD-9-CM 415.1), DVT (ICD-9-CM 451.x and 453.x), hemorrhagic events (Appendix), and acute hepatic events (ICD-9-CM 570.x, 573.3, and 782.4) (where ‘x’ can be any value). The cardiovascular and hepatic endpoints of interest were serious, acute conditions, and were identified from hospital primary diagnoses with the use of previously validated algorithms for VTE [8], ischemic stroke [9,10], and ACS [11]. Serious bleeding endpoints were identified with the use of codes adapted from previous studies [12–15]. Hepatic endpoints included: jaundice (782.4), hepatitis (573.3, unspecified, including toxic hepatitis), and hepatic necrosis and hepatic failure (570.x).

For each cohort, baseline characteristics (e.g. age, gender, and comorbidities) were summarized as the number and proportion of patients. Incidence rates for each outcome were calculated as the number of new cases divided by the person-time at risk, and reported as rates per 1000 person-years with 95% exact binomial confidence intervals (CIs) [16].

Results

Knee replacement surgery

Between 1 July 2004 and 30 June 2008, we identified 143 458 individuals with knee replacement surgery. We applied the following exclusion criteria: < 18 years of age (n = 44); < 180 days of continuous baseline enrollment (n = 17 753); no prescription drug coverage (n = 2817); baseline diagnosis of bleeding or hepatobiliary disease (n = 9159); and baseline treatment with anticoagulants, fibrinolytics, antiplatelet agents, or antifibrinolytics (n = 16 216). After these exclusions, 97 469 knee replacement patients were included in the analysis. As shown in Table 1, the cohort had a higher proportion of women (63%), and half were between the ages of 40 and 64 years. The median duration of anticoagulant use was 13 days (range: 0–364 days), and the median patient follow-up was 70 days (range: 1–365 days). As in the general population, hypertension (53%), obesity or hyperlipidemia (28.5%) and diabetes (18%) were common comorbidities identified in this surgical cohort. Approximately 99% of the individuals were diagnosed with osteoarthritis at some point during the 6 months prior to surgery, suggesting that osteoarthritis was the main reason for the surgery. Forty-four per cent of the knee replacement cohort had no outpatient claims for anticoagulants within 1 week after discharge from the hospital. Among patients prescribed anticoagulants, warfarin was the most commonly used (60%), followed by low molecular weight heparin (34%), fondaparinux (5%), and others (1%).

Table 1.   Baseline characteristics
 Knee replacement cohort*Hip replacement cohort*
  1. NA; SD, standard deviation.

  2. *Includes patients with and without outpatient claims for anticoagulant therapy.

  3. Person-years vary for each outcome because each endpoint is analyzed separately and person-time is truncated at the time of each endpoint.

Persons, n (%)97 469 (100)45 203 (100)
Person-years17 502.1–18 008.78120.5–8332.2
Duration of follow-up (days), median (range)70 (1–365)71 (1–365)
Anticoagulation therapy postdischarge (%)4445
Gender, n (%)
 Male35 729 (36.7)20 261 (44.8)
 Female61 740 (63.3)24 942 (55.2)
Age category (years), n (%)
 18 to < 40414 (0.4)1099 (2.4)
 40 to < 6550 265 (51.6)23 136 (51.2)
 65 to < 7526 251 (26.9)10 000 (22.1)
 ≥ 7520 539 (21.1)10 968 (24.3)
Age (years), mean (SD, median, range)65.1 (10.2, 64, 18–104)64.2 (12.7, 63, 18–102)
Index year, n (%)
 2004–200663 882 (65.5)29 849 (66.0)
 2007–200833 587 (34.5)15 354 (34.0)
Region, n (%)
 Northeast7081 (7.3)4302 (9.5)
 North central38 857 (39.9)16 641 (36.8)
 South35 932 (36.9)15 147 (33.5)
 West15 263 (15.7)8922 (19.7)
 Unknown336 (0.3)191 (0.4)
Comorbidities, n (%)
 Hypertension51 592 (52.9)20 522 (45.4)
 Obesity or hyperlipidemia27 821 (28.5)11 598 (25.7)
 Diabetes17 518 (18.0)5352 (11.8)
 Heart failure3937 (4.0)1774 (3.9)
 Atherosclerosis2384 (2.4)1152 (2.5)
 Peripheral vascular disease3110 (3.2)1514 (3.3)
 Cancer7467 (7.7)3844 (8.5)
Indication for surgery, n (%)
 Osteoarthritis96 068 (98.6)41 654 (92.1)
Surgery type, n (%)
 Primary total94 631 (97.1)41 230 (91.2)
 Primary partial02094 (4.6)
 Revision of implants2838 (2.9)1879 (4.2)

The number of cases, incidence rates and 95% CIs for each endpoint are shown in Table 2. Among our study outcomes, VTE events (n = 1117) were most commonly reported, with an incidence rate of 63.82 per 1000 person-years. Incidence rates for PE and DVT were estimated to be 31.02 per 1000 person-years and 39.17 per 1000 person-years, respectively. The incidence rates for hemorrhagic stroke and ischemic stroke were 14.95 and 0.44 per 1000 person-years, respectively. The incidence rate for ACS was 15.29 per 1000 person-years. Serious bleeds were reported for 826 individuals, resulting in an incidence rate of 46.18 per 1000 person-years. Fifteen cases of acute hepatic events, defined as acute and subacute necrosis of the liver, hepatitis, or jaundice, were identified after knee replacement therapy, resulting in an estimated incidence rate of 0.83 per 1000 person-years.

Table 2.   Incidence rates in knee and hip replacement cohorts
EndpointKnee replacement cohort*Hip replacement cohort*
Person-yearsNumber of casesIncidence rate (95% CI)Person-yearsNumber of casesIncidence rate (95% CI)
  1. CI, confidence interval.

  2. *Includes patients with and without outpatient claims for anticoagulant therapy.

  3. Per 1000 person-years.

  4. Total number of cases for venous thromboembolism includes only the first event. Therefore, it does not equal the sum of pulmonary embolism and deep vein thrombosis, because some patients may have both events.

Acute stroke – hemorrhagic18 008.6780.44 (0.19–0.88)8332.2240.48 (0.13–1.23)
Acute stroke – ischemic17 588.7526314.95 (13.21–16.86)8120.4515318.84 (16.00–22.04)
Acute coronary syndrome17 791.5227215.29 (13.54–17.20)8228.6214617.74 (15.00–20.83)
Venous thromboembolic events17 502.14111763.82 (60.24–67.54)8156.5236744.99 (40.60–49.72)
Pulmonary embolism17 890.0655531.02 (28.53–33.67)8299.4617921.57 (18.55–24.93)
Deep vein thrombosis17 793.3369739.17 (36.37–42.13)8270.2123027.81 (24.37–31.59)
Hemorrhagic events17 886.5382646.18 (43.15–49.36)8278.7838946.99 (42.53–51.77)
Hepatic events18 008.03150.83 (0.47–1.37)8331.28101.2 (0.58–2.21)

Hip replacement surgery

Between 1 July 2004 and 30 June 2008, we identified 67 803 individuals with hip replacement surgery. We applied the following exclusion criteria: < 18 years of age (n = 62); < 180 days of continuous baseline enrollment (n = 8720); no prescription drug coverage (n = 1548); baseline diagnosis of bleeding or hepatobiliary disease (n = 4402); and baseline treatment with anticoagulants, fibrinolytics, antiplatelet agents, or antifibrinolytics (n = 7868). After these exclusions, 45 203 hip replacement patients were included in the analysis. As shown in Table 1, the cohort was predominantly female (55%), and half were between the ages of 40 and 64 years. The median duration of anticoagulant use was 11 days (range: 0–364 days), and the median patient follow-up was 71 days (range: 1–365 days). As in the knee replacement cohort, hypertension (45%), obesity or hyperlipidemia (26%) and diabetes (12%) were common comorbidities identified in this cohort. Approximately 92% of the individuals were diagnosed with osteoarthritis at some point during the 6 months prior to surgery, suggesting that osteoarthritis was the main reason for the surgery. Forty-five per cent of the hip replacement cohort had no outpatient claims for anticoagulants within 1 week after discharge from the hospital. Among patients prescribed anticoagulants, warfarin was most commonly used (59%), followed by low molecular weight heparin (36%), fondaparinux (4%), and others (1%).

The number of cases, incidence rates and 95% CIs for each endpoint are shown in Table 2. Among the study outcomes, serious bleeding events (n = 389) were the most commonly reported among the hip replacement study population, with an estimated incidence rate of 46.99 per 1000 person-years. VTE events were also studied; 367 cases were identified, giving an estimated incidence rate of 44.99 per 1000 person-years. PE was slightly less common than DVT, with incidence rate estimates of 21.57 per 1000 person-years and 27.81 per 1000 person-years, respectively. The incidence rate of hemorrhagic stroke was 0.48 per 1000 person-years, and that for ischemic stroke was 18.84 per 1000 person-years. The estimated incidence rate for ACS was 17.74 per 1000 person-years. Ten cases of acute hepatic events resulted in a low estimated incidence rate of 1.2 per 1000 person-years.

Discussion

There have been few observational studies of thromboembolic events following major orthopedic surgery [17–21]. Differences between studies in the characteristics (e.g. age) of study populations, definitions of outcomes and durations of follow-up complicate comparisons of results across studies. The present study was consistent with a study from five Nordic countries, in that total hip replacements were more common in women than in men [20]. In addition, we observed a higher percentage of women with knee replacements than with hip replacements, consistent with a voluntary national registry (GLORY) in the USA from 1996 to 2000 [21]. In contrast, individuals in the present study were younger than those in the GLORY registry (median ages of 64 and 63 years, as compared with 70 and 68 years for knee replacement and hip replacement, respectively), but older than in a study that used another commercial insurance claims database during the period between 2002 and 2006 [17] (mean age of approximately 65 years, as compared with 55 years).

The present study relied on diagnosis codes reported on claims data, and found incidence rates for VTE of 6.4% for the knee replacement cohort and 4.5% for the hip replacement cohort. In the GLORY registry [21], more stringent criteria were used to define DVT and PE: symptomatic events, subsequently confirmed by diagnostic imaging techniques, including venography, ultrasound, radio-isotope scanning, and three-dimensional computed tomography, led to lower incidence rates; the incidences of DVT were 1.1% (95% CI 0.9–1.4%) at a mean of 22.5 days after total hip replacement surgery and 1.6% (95% CI: 1.4–1.9%) at a mean of 9.5 days after total knee replacement surgery. On the basis of data from insurance claims, Kwong et al. [17] reported 30-day and 90-day incidence rates of 4.9–7.4%, depending on the type of anticoagulant therapy. Oster et al. [18] reported that 7% of study subjects had a VTE diagnosis within 90 days of hospital admission for major knee or hip surgery. Shorr et al. [19] found that the VTE incidence rates ranged from 1.5% to 4.2% for different anticoagulants, with an overall rate of 2.5% during a follow-up period of approximately 2 months. Although the follow-up period in the present study was similar to that in the study by Shorr et al., the definition of VTE was broader, hindering a direct comparison between the studies. However, the VTE incidence rates in the present study appeared to be similar to those in the study by Kwong et al.

For bleeding events, the studies by Kwong et al. and Shorr et al. reported incidence rates for major bleeding of 0.3% [17] and 1.5% [19], respectively, whereas in the present study, the incidence of bleeding events was 4.6%. An explanation for this difference is a broader inclusion of bleeding events in the present study than in the studies by Kwong et al. and Shorr et al. [17,19]. The study by Kwong et al. defined major bleeding according to ICD-9-CM 568.81 (hemoperitoneum), 430.xx (subarachnoid hemorrhage), 431.xx (intracerebral hemorrhage), 432.xx (other and unspecified intracranial hemorrhage), and 998.11 (hemorrhage complicating a procedure). In the study by Shorr et al., a bleeding event was defined by the presence of an ICD-9-CM code for a hemoperitoneal bleed, intracranial hemorrhage or hemorrhagic stroke, hemorrhage complicating a procedure, or any other bleeding diagnosis requiring a transfusion of two or more units of blood as documented on billing records.

We are unaware of any other published observational studies that have assessed the incidence rates for stroke, ACS and hepatic events following orthopedic surgery. In the present study, stroke and ACS had similar incidence rates, and, although the rates were smaller than that of VTE, they constitute important thromboembolic risks, and the incidence rates of these events may be greater than in the general population (about five per 1000 person-years for both stroke and ACS) [22,23]. Although hepatic events in this population were rare, there has been a concern about the hepatic safety profile of anticoagulants, primarily because of the serious liver events that led to a market withdrawal of a direct thrombin inhibitor, ximelagatran [24]. Postsurgical hepatic events could be the result of anesthesia as well as anticoagulants.

Evidence has shown that extending prophylaxis improves clinical outcomes [4], and current practice guidelines now recommend prolonged anticoagulant therapy for 10–35 days after major orthopedic surgery [4]. Despite the recommendation, there remain a significant proportion of individuals who underwent orthopedic surgery but who were not prescribed prophylactic measures. A study of 188 800 surgical discharge records during 2005–2006 found that approximately 20% of surgery patients did not receive VTE prophylaxis during hospitalization [25]. Another study found that only 19% of persons undergoing knee or hip replacement surgery received postdischarge anticoagulant therapy [26]. These findings raise questions about adherence to clinical guidelines, prophylactic measures that may have been used in those who were not prescribed anticoagulants (e.g. aspirin), and the health outcome profile in those who did not use prophylactic measures.

Recently, there has been controversy about differences between guidelines issued by the ACCP [4] and the AAOS [5]. These guidelines weigh risks and benefits differently [27], especially regarding the clinical importance of DVT as a surrogate for PE, and asymptomatic DVT identified from venograms [27–29]. Another difference is that the AAOS guidelines include aspirin therapy. As with other prescription claims databases [26], the current data do not include the use of aspirin obtained over the counter. For these reasons, assessing adherence to clinical guidelines is not straightforward. This study measured postsurgical adverse event rates in a large US database, but did not attempt to link adverse events to antiplatelet or anticoagulant use. Studies such as this one, which quantify the risks of clinical endpoints related to bleeding and thromboembolism, will help to inform this debate. In addition to further observational studies, large randomized clinical trials will be important in understanding the effects of these medications on the risks of events such as stroke and ACS.

Among the strengths of the present study was the large sample size, yielding better precision of incidence estimates. This study shares limitations with other database studies, in that the study utilized medical insurance claims, which were not designed for research purposes. We used case-finding algorithms that have been validated by other investigators, but we did not validate claims diagnoses in this study. If endpoints were coded incorrectly to other diagnoses, or if they were coded as secondary rather than primary diagnoses, they would not have been included as endpoints in this study. Conversely, if patients were coded as having primary diagnosis endpoints that did not occur (e.g. as a rule-out diagnosis), such patients could have been misclassified as cases in this study.

In conclusion, patients who underwent hip or knee replacement surgery appeared to have comorbidities that are similar to those in the general population, except for a predominantly high proportion of osteoarthritis, for which the surgery was probably indicated. In comparing incidence rates for outcomes across different studies, one should consider patient characteristics, definitions of treatments and outcomes, and durations of follow-up across studies. The results reported here could help in the making of challenging decisions regarding the clinical management of risks attributable to bleeding events and clotting events.

Disclosure of Conflict of Interests

This study was sponsored by Bristol-Myers Squibb. S. Lanes and K. Fraeman are employees of United BioSource Corporation, which received a research fund from Bristol-Myers Squibb. A. Myers, J. Wood Ives, and H.Y. Huang are employees of Bristol-Myers Squibb.

Appendix

ICD-9-CM codeDescription
078.6Hemorrhagic nephrosonephritis
246.3Hemorrhage and infarction of thyroid
286.5Hemorrhagic disorder caused by intrinsic circulating anticoagulants
360.43Hemophthalmos, except current injury
362.43Hemorrhagic detachment of retinal pigment epithelium
362.81Retinal hemorrhage
363.61Unspecified choroidal hemorrhage
363.62Expulsive choroidal hemorrhage
363.72Hemorrhagic choroidal detachment
364.41Hyphema
372.72Conjunctival hemorrhage
374.81Hemorrhage of eyelid
376.32Orbital hemorrhage
377.42Hemorrhage in optic nerve sheaths
379.23Vitreous hemorrhage
423Hemopericardium
423.0Hemopericardium
430Subarachnoid hemorrhage
431Intracerebral hemorrhage
432Other and unspecified intracranial hemorrhage
432.0Intracranial hemorrhage
432.1Subdural hemorrhage/intracranial hemorrhage
432.9Unspecified intracranial hemorrhage
456Esophageal varices with bleeding
456.0Esophageal varices with bleeding
456.20Esophageal varices with bleeding in diseases classified elsewhere
459Unspecified hemorrhage
459.0Unspecified hemorrhage
530.82Esophageal hemorrhage
531Acute gastric ulcer with hemorrhage
531.0Acute gastric ulcer with hemorrhage
531.00Acute gastric ulcer with hemorrhage, without mention of obstruction
531.01Gastrointestinal/acute stomach ulcer with hemorrhage and obstruction
531.2Acute gastric ulcer with hemorrhage and perforation
531.20Acute gastric ulcer with hemorrhage and perforation, without mention of obstruction
531.21Gastrointestinal/acute stomach ulcer with hemorrhage/perforation and obstruction
531.4Chronic or unspecified gastric ulcer with hemorrhage
531.40Chronic or unspecified gastric ulcer with hemorrhage, without mention of obstruction
531.41Gastrointestinal/chronic stomach ulcer with hemorrhage and obstruction
531.6Chronic or unspecified gastric ulcer with hemorrhage and perforation
531.60Chronic or unspecified gastric ulcer with hemorrhage and perforation, without mention of obstruction
531.61Gastrointestinal/chronic stomach ulcer with hemorrhageperforation and obstruction
532Acute duodenal ulcer with hemorrhage
532.0Acute duodenal ulcer with hemorrhage
532.00Acute duodenal ulcer with hemorrhage, without mention of obstruction
532.01Gastrointestinal/ac duodenal ulcer with hemorrhage and obstruction
532.2Acute duodenal ulcer with hemorrhage and perforation
532.20Acute duodenal ulcer with hemorrhage and perforation, without mention of obstruction
532.21Acute duodenal ulcer with hemorrhage, perforation, and obstruction
532.4Chronic or unspecified duodenal ulcer with hemorrhage
532.40Chronic or unspecified duodenal ulcer with hemorrhage, without mention of obstruction
532.41Chronic or unspecified duodenal ulcer with hemorrhage and obstruction
532.6Chronic or unspecified duodenal ulcer with hemorrhage and perforation
532.61Chronic or unspecified duodenal ulcer with hemorrhage, perforation, and obstruction
533Acute peptic ulcer, unspecified site, with hemorrhage
533.0Acute peptic ulcer, unspecified site, with hemorrhage
533.00Acute peptic ulcer, unspecified site, with hemorrhage, without mention of obstruction
533.01Acute peptic ulcer, unspecified site, with hemorrhage and obstruction
533.2Acute peptic ulcer, unspecified site, with hemorrhage and perforation
533.20Acute peptic ulcer, unspecified site, with hemorrhage and perforation, without mention of obstruction
533.21Acute peptic ulcer, unspecified site, with hemorrhage, perforation, and obstruction
533.4Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage
533.40Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage, without mention of obstruction
533.41Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage and obstruction
533.6Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage and perforation
533.60Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage and perforation, without mention of obstruction
533.61Chronic or unspecified peptic ulcer, unspecified site, with hemorrhage, perforation, and obstruction
534Acute marginal ulcer with hemorrhage
534.0Acute gastrojejunal ulcer with hemorrhage
534.00Acute gastrojejunal ulcer with hemorrhage, without mention of obstruction
534.01Acute gastrojejunal ulcer with hemorrhage and obstruction
534.2Acute gastrojejunal ulcer with hemorrhage and perforation
534.20Acute gastrojejunal ulcer with hemorrhage and perforation, without mention of obstruction
534.21Acute gastrojejunal ulcer with hemorrhage, perforation, and obstruction
534.4Chronic or unspecified gastrojejunal ulcer with hemorrhage
534.40Chronic or unspecified gastrojejunal ulcer with hemorrhage, without mention of obstruction
534.41Chronic or unspecified gastrojejunal ulcer with hemorrhage and obstruction
534.6Chronic or unspecified gastrojejunal ulcer with hemorrhage and perforation
534.60Chronic or unspecified gastrojejunal ulcer with hemorrhage and perforation, without mention of obstruction
534.61Chronic or unspecified gastrojejunal ulcer with hemorrhage, perforation, and obstruction
535.01Acute gastritis with hemorrhage
535.11Atrophic gastritis with hemorrhage
535.21Gastric mucosal hypertrophy with hemorrhage
535.31Alcoholic gastritis with hemorrhage
535.41Other specified gastritis with hemorrhage
535.51Unspecified gastritis and gastroduodenitis with hemorrhage
535.61Duodenitis with hemorrhage
537.83Angiodysplasia of stomach and duodenum with hemorrhage
562.02Diverticulosis of small intestine with hemorrhage
562.03Diverticulitis of small intestine with hemorrhage
562.12Diverticulosis of colon with hemorrhage
562.13Diverticulitis of colon with hemorrhage
568.81Hemoperitoneum (non-traumatic)
569.3Hemorrhage of rectum and anus
569.85Angiodysplasia of intestine with hemorrhage
578Gastrointestinal hemorrhage
578.0Hematemesis
578.1Blood in stool
578.9Unspecified hemorrhage of gastrointestinal tract
596.7Hemorrhage into bladder wall
599.7Hematuria
602.1Congestion or hemorrhage of prostate
620.1Corpus luteum cyst or hematoma
621.4Hematometra
626.2Excessive or frequent menstruation
626.3Pubertal bleeding
626.5Ovulatory bleeding
626.6Metrorrhagia
626.7Postcoital bleeding
626.8Other disorder of menstruation and other abnormal bleeding from female genital tract
626.9Unspecified disorder of menstruation and other abnormal bleeding from female genital tract
627.0Premenopausal menorrhagia
627.1Postmenopausal bleeding
627.4Symptomatic states associated with artificial menopause
640Hemorrhage in early pregnancy
641.9Unspecified antepartum hemorrhage
666.1Other immediate postpartum hemorrhage
719.1Hemarthrosis
719.11Hemarthrosis – shoulder
719.12Hemarthrosis – upper arm
719.13Hemarthrosis – forearm
719.14Hemarthrosis – hand
719.15Hemarthrosis – pelvis
719.16Hemarthrosis – lower leg
719.17Hemarthrosis – ankle
719.18Hemarthrosis – other
719.19Hemarthrosis – multiple sites
782.7Spontaneous ecchymoses
784.7Epistaxis
784.8Hemorrhage from throat
786.3Hemoptysis
852Subarachnoid, subdural and extradural hemorrhage following injury
852.0Subarachnoid hemorrhage following injury without mention of open intracranial wound
852.2Subdural hemorrhage following injury without mention of open intracranial wound
852.4Extradural hemorrhage following injury without mention of open intracranial wound
853.0Intracranial hemorrhage without mention of open intracranial wound
958.2Secondary and recurrent hemorrhage as an early complication of trauma
997.02Iatrogenic cerebrovascular infarction or hemorrhage
998.11Hemorrhage complicating a procedure
E934.2External agents affecting blood causing adverse effects – anticoagulants*
E934.4External agents affecting blood causing adverse effects – fibrinolytics*
E934.5External agents affecting blood causing adverse effects – anticoagulant antagonists and other coagulants*
E934.9External agents affecting blood causing adverse effects – unspecified*

Ancillary