All patients discharged between 2001 and 2010 with the
diagnosis of CSM who were 25 years or older were included.
Patients had undergone anterior and/or posterior cervical
fusion or laminoplasty at C2 or below. Patients with a history
of cancer, vertebral fracture or fusion of 9+ vertebrae were
Demographic data included age, gender, race and primary
payer. The comorbidity burden was calculated using the Deyo
comorbidity index, an adapted version of the Charleston
comorbidity index. Hospital course data included admission
type, hospital size, teaching status and location, length of stay,
discharge status and total charges. In hospital complications
and mortality were determined. Patients of different ages were
assessed with two dichotomous comparisons: 25–64 years old
compared to 65+ years old and 65–75 versus 76+ years old.
In total, 54,416 patients were identified. This included
35,319 in the 25–64 year-old group and 19,097 in the 65+
year-old group. As could be expected, the Deyo comorbidity index was significantly higher for the 65+ group (0.79 vs
0.44, p<0.0001). The hospital course for patients in the 65+
group was longer, 4.76 vs 3. 26 days (p<0.0001). These older
patients had significantly higher procedure-related morbidity
( 11.39% vs 5.93%, p<0.0001). Factors that were significantly
higher for the older group included cardiovascular events,
respiratory complications (including acute respiratory distress
syndrome), urinary complications, post-hemorrhagic anemia
and hematoma/seroma formation. The 65+ group also had a
higher mortality rate (.85% vs .2%, p<.001). Despite the many
increased risks associated with the 65+ group, the 25–64 group
did have a higher incidence of device-related complications
( 1.52% vs 0.91%, p<0.001).
The NIS database was used to identify 35,319 patients aged
25-64 and 19,097 patients aged 65+ who were treated with
anterior and/or posterior cervical fusion or laminoplasty for
cervical myelopathy from 2001-2010.
Procedure-related complications were compared between
the two groups. Increased age was shown to be associated
with increased preoperative comorbidities, increased
procedure-related morbidity and increased length/cost of
The 65+ year old group had higher procedure-related
morbidities in every category when compared to the 25-64
group except for device-related complications, even when
trying to control for baseline comorbidities.
When the 65+ group was further stratified into 65-75 and
76+ groups, age-related comorbidities and procedure-related morbidities again increased.
Strengths of Study:
This is a national database study that includes a large cohort.
The topic is relevant. The presented analysis draws attention
to a specific variable—age—that can be used to risk stratify
patients undergoing cervical spine procedures.
The model accuracy was assessed using cross-validation.
Limitations of Study:
There are limitations with all national database studies. This
is particularly true of the NIS database which is based on
administratively coded data and where there is no follow-up
after discharge from the hospital.
Although comorbidities were attempted to be controlled for,
this can be challenging. There is certainly a chance that some
of the differences were at least partially related to underlying
comorbidities and not just related to age.
The NIS database does not allow for spine-specific clinical
outcomes data. It would have been useful to compare the
clinical outcomes between the two groups in addition to the
morbidity and mortality.
On further analysis, the 65+ year-old patients were more
likely to be discharged to other hospital facilities ( 32.28% vs
9.89%, p<0.0001) and were more likely to require home health
( 10.8% vs 6. 8 %, p<0.0001). In addition, hospital charges were
significantly higher in the 65+ when compared to the 25–64
group ($57,449.94 vs $49,951.11, p<0.0001).
In a secondary dichotomous comparison, the older patients were stratified into two cohorts, 65–75 years old and
76+ years old. Again, as one might expect, the older group
showed higher preoperative comorbidities and postoperative
complications, although incrementally less than the previous comparison. The Deyo comorbidity score was 0.77 in the
65–75 group and 0.83 for the 76+ group (p<0.0001). In addition, they have an increased total complication rate of 13.87%
vs 10.2% (p<0.001). The three postoperative complications
with statistically significant increases in the 76+ group were
shock ( 6. 34[ 11. 16-3.60]), digestive system ( 1.92[ 2. 40-1.54])
and wound dehiscence ( 1.71[ 2.56-1.15]). These values were
reported as odds ratios with the 95% confidence interval. The
mortality risk in the 76+ group was 1.53% which was significantly higher than the 65–75 group at 0.85% ( 2.60[ 3.05-2.22]).
Overall, this study evaluates the impact of advanced age as
an independent predictor of inpatient outcome measures after
fusion or laminoplasty surgery for CSM. The effect of age on
surgical outcomes and complications has been investigated
for a number of surgical pathologies, but this study adds a
large patient cohort analysis utilizing a national database.
As one might have expected, this study showed a higher rate
of perioperative complications with increasing patient age
in the CSM population. This includes mortality, even when