Surgical Endoscopy

, Volume 27, Issue 11, pp 4038–4043

Can laparoscopy for colon resection reduce the need for discharge to skilled care facility?

Authors

  • Abhijit Shaligram
    • Department of SurgeryUniversity of Nebraska Medical Center
  • Lynette Smith
    • Department of BiostatisticsUniversity of Nebraska Medical Center
  • Pradeep Pallati
    • Department of SurgeryUniversity of Nebraska Medical Center
  • Anton Simorov
    • Department of SurgeryUniversity of Nebraska Medical Center
  • Jane Meza
    • Department of BiostatisticsUniversity of Nebraska Medical Center
    • Department of SurgeryUniversity of Nebraska Medical Center
Article

DOI: 10.1007/s00464-013-3052-y

Cite this article as:
Shaligram, A., Smith, L., Pallati, P. et al. Surg Endosc (2013) 27: 4038. doi:10.1007/s00464-013-3052-y

Abstract

Background

A significant proportion of patients, especially the elderly undergoing colon resections, are likely to be discharged to a skilled care facility. This study aims to examine whether the technique of colectomy, open versus laparoscopic, contributed to their discharge to a skilled care facility.

Methods

This was a retrospective analysis using discharge data from the Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project, Agency for Healthcare Research and Quality. Adult patients who underwent colectomy in 2009 were evaluated. SAS and SUDAAN software were used to provide weighted estimates and to account for the complex sampling design of the NIS. We compared routine discharge to nonroutine discharge, defined as transfer to short-term hospital, skilled nursing facility, intermediate care, home health, or another type of facility.

Results

A weighted total of 221,294 adult patients underwent colectomy in 2009 and had the primary outcome of discharge available. Of these colon resections, 70,361 (32 %) were performed laparoscopically and 150,933 (68 %) by open technique. A total of 139,047 (62.8 %) patients had routine discharge and 73,572 (33.3 %) nonroutine. A total of 8,445 (3.8 %) patients died while in the hospital, and 229 (0.1 %) left against medical advice and were excluded from further analysis. On univariate analysis, age ≥65 years, female gender, Black/Hispanic race, open technique (compared to laparoscopic), Medicare/Medicaid insurance status, comorbidity index of ≥1, and malignant primary diagnosis predicted nonroutine discharge. A multivariate logistic model was then used to predict nonroutine discharge in these patients using variables significant in the univariate analysis at the α = 0.05 significance level. In the multivariate analysis, open compared to laparoscopic technique was independently associated with increased likelihood of discharge to skilled care facilities (odds ratio 2.85, 95 % confidence interval 2.59–3.14).

Conclusions

In addition to the expected factors like advancing age, female gender, and increasing comorbidity index, open compared to laparoscopic technique for colectomy is associated with an increased likelihood of discharge to skilled care facilities. When feasible, the laparoscopic technique should be considered as an option, especially in the elderly patients who require colon resection, because it may reduce their likelihood of discharge to a skilled care facility.

Keywords

BowelGut

Laparoscopic technique for colon resection was first described over 20 years ago [1]. Several studies over the past two decades have established the safety of laparoscopic colectomy for several pathologic conditions [2, 3]. Many reports have shown the short-term benefit of laparoscopy colectomy for patients with regard to shorter hospital stay, early return to full activities, and fewer wound-related complications [46]. There has been a trend toward increased utilization of laparoscopic approach for colon resections, with now over a third of cases being done by this technique [7, 8].

Several studies have focused on identifying the predictors of mortality and morbidity [912]. Most factors focus on patient characteristics like age, gender, poor preoperative exercise tolerance, American Society of Anesthesiologists classification, comorbidities including dyspnea, ascites, congestive heart failure, dialysis, or disseminated cancer, or postoperative conditions including reintubation, renal failure, stroke, and septic shock [9, 10]. Understanding these factors is of critical importance, but most of them cannot be altered. With the introduction of a laparoscopic approach to surgery, this is one instance where the surgeon and the patient can make a choice that could affect outcome.

Colon resection is a major surgical event, especially for an older person. Often, it is followed by a marked decrease in the level of functioning a patient is able to achieve after surgery [13]. It is even more debilitating when a previously independent person must be discharged to an institution. As the number of elderly patients undergoing colon resection has increased, a significant proportion of patients are likely to be discharged to a skilled care facility. Whereas most studies have focused on the outcomes of laparoscopic colectomy, few have looked at the impact of surgical technique on discharge disposition. This study aims to examine whether the technique of colectomy, open versus laparoscopic, contributed to the discharge of patients to the skilled care facility.

Methods

Data source

Data were obtained from the Nationwide Inpatient Sample (NIS), a database developed as part of the Healthcare Cost and Utilization Project, sponsored by the Agency for Healthcare Research and Quality. The NIS is designed to approximate a 20 % sample of US community hospitals.

The NIS 2009 database was queried to identify patients with an International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), primary procedure codes for colectomy. The procedure codes for open colectomy were 45.71, 45.72, 45.73, 45.74, 45.75, 45.76, 45.82, and 45.79, and for laparoscopic colectomy were 17.31, 17.32, 17.33, 17.34, 17.35, 17.36, 17.39, and 45.81.

Comorbidity assessment

Patient comorbidities were assessed by the Romano modification of the Charlson comorbidity score. The weighted comorbidities were summed to create a comorbidity index for each person in the sample. The index was further categorized as 0 (no preexisting weighted comorbidity), 1 (1 weighted comorbid condition), 2 (2 weighted comorbid conditions), and ≥3 (3 or >3 weighted comorbid conditions).

Outcome measures

Disposition of patient included the following: (1) routine, (2) transfer to short-term hospital, (5) other transfers, including skilled nursing facility, intermediate care, and another type of facility, (6) home health care, (7) against medical advice, (20) died in hospital, and (99) discharged alive, destination unknown. We compared (1) routine disposition to (2, 5, 6) nonroutine: transfer to short-term hospital, other transfers, including skilled nursing facility, intermediate care, and another type of facility, home health care.

Statistical methods

SAS software (SAS Institute Inc., Cary, NC, USA) and SUDAAN software were used for all statistical analysis to account for the complex sampling design of NIS. Weighted sample estimates, standard errors, and 95 % limits were calculated by the Taylor expansion method. χ2 tests were used to compare patient characteristics by discharge disposition. A multivariate logistic model was used to predict nonroutine disposition in patients who underwent colectomy using variables significant in the univariate analysis at the α = 0.20 significance level. All statistical tests were two-sided, and p values of <0.05 were considered statistically significant.

Results

A weighted total of 221,294 adult patients underwent colectomy in 2009 and had the primary outcome of discharge available. Of these colon resections, 70,361 (32 %) were performed laparoscopically and 150,933 (68 %) by open technique. A total of 139,047 (62.8 %) patients had routine discharge and 73,572 (33.3 %) nonroutine. A total of 8,445 (3.8 %) patients died while in the hospital, and 229 (0.1 %) left against medical advice and were excluded from further analysis. The most common comorbid conditions in this patient population included chronic obstructive airway disease (15.6 %), diabetes mellitus (15.6 %), congestive heart failure (6.8 %), peripheral vascular disease (4.9 %), and history of myocardial infarction (4 %). Discharge disposition by procedure type is listed in Table 1.
Table 1

Discharge disposition by procedure type

Total

Total

Laparoscopic

Open

n

%

n

%

n

%

Total

221294

100

70361

100

150933

100

Routine

139047

62.8

57831

82.2

81216

53.8

Other facility

35256

15.9

4499

6.4

30757

20.4

Home health

38316

17.3

7461

10.6

30855

20.4

Against medical advice

229

0.1

33

0.05

197

0.13

Died

8445

3.8

537

0.8

7908

5.2

On univariate analysis, age ≥65 years, female gender, Black/Hispanic race, open technique (compared to laparoscopic), Medicare/Medicaid insurance status, comorbidity index of ≥1, and diagnosis (e.g., hemorrhage, malignancy, inflammatory bowel disease) predicted nonroutine discharge (Table 2). In the multivariate analysis, open compared to laparoscopic technique was independently associated with increased likelihood of discharge to skilled care facilities (odds ratio 2.85, 95 % confidence interval 2.59–3.14) (Table 3).
Table 2

Patient and hospital characteristics of those who underwent colectomy with univariate analysis of nonroutine discharge complication

Characteristics

Variables

Overall weighted

Routine discharge

Nonroutine discharge

p value

n

%

n

%

n

%

Total

 

212619

100

139047

65.4

73572

34.6

 

Age

<65 years

104065

49.3

81228

78.1

22837

22.0

<0.0001

≥65 years

106935

50.7

22837

22.0

50462

47.2

 

Gender

Male

99555

46.9

67814

68.1

31741

31.9

<0.0001

Female

112651

53.1

70830

62.9

41821

37.1

 

Race

White

144037

67.7

93159

64.7

50878

35.3

0.016

Black

15863

7.5

10463

66.0

5400

34.0

 

Hispanic

12945

6.1

9001

69.5

3945

30.5

 

Other/unknown

39774

18.7

26424

66.4

13350

33.6

 

Surgical technique

Laparoscopic

69791

32.8

57831

82.9

11960

17.1

<0.0001

Open

142828

67.2

81216

56.9

61612

43.1

 

Insurance status

Medicare/Medicaid

113262

53.3

60223

53.2

53039

46.8

<0.0001

Private/self-pay/other

99358

46.7

78824

79.3

20533

20.7

 

Comorbidity index

0 Comorbid conditions

78739

37.0

58241

74.0

20497

26.0

<0.0001

1 Comorbid condition

15108

7.1

9089

60.2

6019

39.8

 

2 Comorbid conditions

46245

21.8

32044

69.3

14201

30.7

 

3+ Comorbid conditions

72527

34.1

39672

54.7

32855

45.3

 

Length of stay

≤10 days

154469

72.7

120411

78.0

34058

22.1

<0.0001

>10 days

58150

27.4

18636

32.1

39514

68.0

 

Bed size of hospital

Small

23202

11.1

15176

65.4

8026

34.6

0.73

Medium

50915

24.3

33596

66.0

17319

34.0

 

Large

135168

64.6

87800

65.0

47367

35.0

 

Hospital location

Rural

24564

11.7

16115

65.6

8449

34.4

0.77

Urban

184721

88.3

120457

65.2

64264

34.8

 

Hospital teaching status

Nonteaching

115560

55.2

75464

65.3

40096

34.7

0.93

Teaching

93724

44.8

61107

65.2

32617

34.8

 

Admission type

Emergency/urgent

81863

42.6

38514

47.0

43349

53.0

<0.0001

Elective

110259

57.4

86214

78.2

24045

21.8

 

Primary diagnosis

Hemorrhage

516

0.2

211

40.8

305

59.2

<0.0001

Inflammatory bowel disease

5392

2.5

2330

43.2

3062

56.8

 

Benign neoplasm

23617

11.1

20310

86.0

3307

14.0

 

Malignancy

73725

34.7

49354

66.9

24371

33.1

 

Infection (gastroenteritis)

506

0.2

324

64.0

182

36.0

 

Obstruction

12022

5.7

6703

55.8

5319

44.2

 

Diverticulitis

58943

27.7

40975

69.5

17968

30.5

 

Constipation

549

0.3

424

77.2

125

22.8

 

Other colon diagnosis

5090

2.4

1660

32.6

3430

67.4

 

Other

32260

15.2

16757

51.9

15503

48.1

 
Table 3

Multivariate analysis of nonroutine discharge of patients who underwent colectomy

Characteristics

Variables

OR

Lower 95 % CI

Upper 95 % CI

p value

Age

<65 years

Ref.

 

≥65 years

3.81

3.43

4.25

<0.0001

Gender

Male

Ref.

 

Female

1.33

1.24

1.43

<0.0001

Race

White

Ref.

 

Black

0.88

0.76

1.02

0.0001

Hispanic

0.62

0.50

0.76

 

Other/unknown

0.96

0.84

1.08

 

Surgical technique

Laparoscopic

Ref.

 

Open

2.12

1.90

2.37

<0.0001

Insurance status

Private/self-pay/other/unknown

Ref.

 

Medicare/Medicaid

2.89

2.58

3.23

<0.0001

Comorbidity index

0 Comorbid conditions

Ref.

 

1–2 Comorbid conditions

1.49

1.35

1.63

<0.0001

3+ Comorbid conditions

2.40

2.17

2.65

 

Admission type

Elective

Ref.

 

Emergency/urgent

3.29

3.02

3.60

<0.0001

Primary diagnosis

Secondary colon diagnosis

Ref.

 

Hemorrhage

0.51

0.29

0.89

 

Inflammatory bowel disease

0.70

0.53

0.91

<0.0001

Benign neoplasm

0.33

0.28

0.39

 

Malignancy

0.41

0.37

0.45

 

Infection (gastroenteritis)

0.70

0.37

1.33

 

Obstruction

0.77

0.68

0.88

 

Diverticulitis

0.52

0.47

0.57

 

Constipation

0.18

0.04

0.75

 

Other primary colon diagnosis

1.20

0.99

1.44

 

OR odds ratio, CI confidence interval

Discussion

During the study period, approximately one-third of the patients undergoing colon resection were either discharged to a skilled facility or required home health. Overall, the rate was 17 % in the laparoscopic group and 40 % in the open group. Laparoscopy was significantly associated with younger patients, those with private/self-pay/other insurance, fewer comorbid conditions, shorter length of stay, medium/large hospitals, urban hospitals, teaching hospitals, elective admission, and primary diagnosis (Table 4). Although some of this could be attributable to the patient selection for the type of procedure and comorbidities, multivariate analysis revealed that surgical technique was an independent risk factor.
Table 4

Patient and hospital characteristics of those who underwent colectomy with univariate analysis of surgical technique

Characteristics

Variables

Overall weighted

Laparoscopic

Open

p value 

n

%

n

%

n

%

Total

 

212619

100

69791

32.8

142828

67.2

 

Age

<65 years

104065

49.3

37357

35.9

66708

64.1

<0.0001

≥65 years

106935

50.7

32007

29.9

74928

70.1

 

Gender

Male

99555

46.9

32797

32.9

66758

67.1

0.46

Female

112651

53.1

36712

32.6

75940

67.4

 

Race

White

144037

67.7

48117

33.4

95921

66.6

0.055

Black

15863

7.5

4761

30.0

11102

70.0

 

Hispanic

12945

6.1

4439

34.3

8507

65.7

 

Other/unknown

39774

18.7

12475

31.4

27299

68.6

 

Insurance status

Medicare/Medicaid

113262

53.3

31948

28.2

81314

71.8

<0.0001

Private/self-pay/other

99358

46.7

37843

38.1

61514

61.9

 

Comorbidity index

0 Comorbid conditions

78739

37.0

30427

38.6

48312

61.4

<0.0001

1 Comorbid condition

15108

7.1

4761

31.5

10348

68.5

 

2 Comorbid conditions

46245

21.8

16160

34.9

30085

65.1

 

3+ Comorbid conditions

72527

34.1

18444

25.4

54083

74.6

 

Length of stay

≤10 days

154469

72.7

61872

40.1

92597

59.9

<0.0001

>10 days

58150

27.4

7919

13.6

50231

86.4

 

Bed size of hospital

Small

23202

11.1

6469

27.9

16734

72.1

0.027

Medium

50915

24.3

17478

34.3

33437

65.7

 

Large

135168

64.6

45025

33.3

90142

66.7

 

Hospital location

Rural

24564

11.7

4856

19.8

19708

80.2

<0.0001

Urban

184721

88.3

64116

34.7

120605

65.3

 

Hospital teaching status

Nonteaching

115560

55.2

34238

29.6

81322

70.4

0.0001

Teaching

93724

44.8

34734

37.1

58991

62.9

 

Admission type

Emergency/urgent

81863

42.6

12853

15.7

69010

84.3

<0.0001

Elective

110259

57.4

50464

45.8

59795

54.2

 

Primary diagnosis

Hemorrhage

516

0.2

98

18.9

418

81.1

<0.0001

Inflammatory bowel disease

5392

2.5

1882

34.9

3510

65.1

 

Benign neoplasm

23617

11.1

14361

60.8

9256

39.2

 

Malignancy

73725

34.7

24368

33.1

49357

67.0

 

Infection (gastroenteritis)

506

0.2

110

21.8

396

78.2

 

Obstruction

12022

5.7

1599

13.3

10423

86.7

 

Diverticulitis

58943

27.7

21450

36.4

37494

63.6

 

Constipation

549

0.3

265

48.3

284

51.7

 

Other colon diagnosis

5090

2.4

498

9.8

4592

90.2

 

Other

32260

15.2

5160

16.0

27100

84.0

 

A number of studies have laid out the predictors of mortality in patients undergoing colectomy [14, 15]. A recent study by Masoomi et al. [15] reported that in patients undergoing colorectal surgery, emergent surgery, liver disease, total colectomy, age ≥65 years, chronic renal failure, and malignant tumor are major risk factors for in-hospital mortality.

The decision to enter long-term care is often a major life event for the elderly and is made in the context of individual circumstances. Many experience premature postoperative functional decline, limiting their ability to return directly to their homes after surgery. Patients discharged to a skilled nursing facility are much more likely to die within the first perioperative year, and skilled nursing facility disposition should be considered as either a marker of debility and or a component of patient decline [13]. Although there are numerous factors like age or comorbidity that cannot be changed, surgical technique is one of the modifiable risk factors where there are alternative approaches available. Often there is only a small amount of difference in the surgical insult required to decrease the level of functioning in an already fragile individual, putting him or her into a nursing home.

Interestingly, our data revealed that hospital size and teaching status or location did not predict discharge to a nonroutine facility. Patients with inflammatory bowel disease were four times more likely to be discharged to a nursing facility or to require home health care. This could be due to the need for home health for the care of ostomy for this subgroup. Insurance status did make a difference, with Medicare/Medicaid patients being more likely to be discharged to a skilled care facility. Although the reason for this finding is not entirely clear, we speculate that this could be a reflection of the easy disposition to get the patients out of the hospital.

The limitations of this study include those inherent to any large administrative database, although the NIS is widely used and has been well validated. The coding errors for diagnoses range from 0.04 to 0.08 % [16]. We have included the discharge disposition to other facility, which includes skilled nursing facility, intermediate care, and another type of facility and home health in one group. We also acknowledge that home health care can have a very wide meaning, from assistance with daily living to minor wound care. However, when we performed analyses excluding the data from the home health group, our findings were similar to when both groups were combined. We think that home health also needs an additional level of care and thus ought to be included.

Laparoscopic colectomy, when feasible, is now regarded as a standard of care. Although it is now more likely to be offered to young patients, because it shows a reduction in length of stay and in complications, it would be prudent to be aggressive in considering this approach in the elderly because it could make a difference in their postsurgical disposition.

Conclusions

In addition to the expected factors like advancing age, female gender, and increasing comorbidity index, open compared to laparoscopic technique for colectomy is associated with an increased likelihood of discharge to skilled care facilities. When feasible, the laparoscopic technique should be considered as an option, especially in elderly patients who require colon resection, because it may reduce their likelihood of discharge to skilled care facilities.

Disclosures

Abhijit Shaligram, Lynette Smith, Pradeep Pallati, Anton Simorov, Jane Meza, and Dmitry Oleynikov have no conflicts of interest or financial ties to disclose.

Copyright information

© Springer Science+Business Media New York 2013