Introduction
During the last decade, various minimally invasive spine surgery (MISS) techniques
have been developed with the aim to reduce approach-related soft-tissue trauma and
its associated complications while achieving the clinical and radiologic outcomes
of open techniques.[1]
[2]
[3]
[4] The MISS quickly proved as an alternative for a short-segment spinal pathology.
With the evolution of modern MISS instrumentations, the possibilities for reduction,
distraction, compression, and reclining on multiple levels increased substantially
and MISS became a therapeutic option even for complex spinal pathologies, traumas,
and deformities. However, there is still controversy regarding some possible disadvantages
of MISS techniques, such as longer operating times, higher intraoperative radiation,
a challenging learning curve,[2]
[3]
[5]
[6]
[7] and a potentially higher risk of cage and pedicle screw misplacements.[5]
The aim of this study is to analyze our institutional experience and personal impressions
using MISS to describe our learning curve and how gaining experience influences different
parameters of the surgical procedure.
Methods
This prospective study included 152 consecutive patients who underwent MISS performed
by a single senior surgeon between April 2013 and December 2018. According to the
protocol, patient demographics, type of pathology, type and number of implants used,
blood loss, duration of surgery, radiation exposure, and length of hospital stay (LOS)
were recorded. The outcome was measured by pre- and postoperative Visual Analog Scale
(VAS) score and Oswestry Disability Index (ODI) questionnaire.
The cohort was divided into consecutive quarters (38 patients each). A comparison
of the results for each quarter and timeline analysis was made to assess the learning
curve of the MISS techniques and how the parameters of the surgical procedure changed
over the time period. A Mann–Whitney U test was used to compare the parameters of the surgical procedure for the different
quarters. The data were analyzed using the computer software IBM SPSS Statistics version
22.0.
A comparison between our institutional experience using MISS and the results from
the literature review was made.
Results
In this series, 152 patients, comprising 70 women (46%) and 82 men (54%), were operated
on. The average patient age at the time of surgery was 49 years (range: 17–81 years).
The average follow-up was 1 year. The etiology of the pathology was traumatic in 65
cases, degenerative in 71 cases, oncologic in 8 cases, and infectious in 8 cases.
Twenty-four patients were obese (body mass index [BMI] > 30), and in 15 cases MISS
was used for revision surgery. Treated levels ranged from Th5 to S1. The number of
treated vertebrae ranged between two and six ([Table 1]).
Table 1
Spinal levels treated and number of segments included
|
Thoracic level
|
14 patients
|
|
Lumbar level
|
103 patients
|
|
TL junction (T11–L2)
|
35 patients
|
|
Short segment (4 screws)
|
67 patients
|
|
Short segment (6 screws)
|
49 patients
|
|
Short segment (5 screws)
|
3 patients
|
|
Long segment stabilizations (≥4 levels)
|
33 patients
|
Only percutaneous transpedicular screw fixation was performed in 65 cases, minimally
invasive transforaminal lumbar interbody fusion (MI-TLIF) in 70 cases and vertebral
body replacement in 4 cases. Augmented screws were used in 13 cases. In total, 817
screws and 86 cages were implanted.
The operative time ranged from 60 to 180 minutes, with an average of 141.6 minutes
for the series ([Table 2]). The difference in the operative time between the first and third quarters was
statistically significant (p = 0.0268), as was for the first and last quarters (p = 0.0136).
Table 2
Duration of surgery for each quarter and for the whole series
|
1st quarter
|
2nd quarter
|
3rd quarter
|
4th quarter
|
All (n = 152)
|
|
155.0 min
|
143.2 min
|
134.5 min
|
133.8 min
|
141.6 min
|
The average radiation exposure time was 70.1 seconds (range: 30–121 seconds) or 6.8
mGy (range: 3.0–12.2 mGy; [Table 3]). Mean time in minutes per screw was 20.8 minutes for procedures with transpedicular
stabilization only. Mean radiation exposure was 14.1 seconds or 1.4 mGy per screw.
Table 3
Radiation exposure for each quarter and for the whole series
|
1st quarter
|
2nd quarter
|
3rd quarter
|
4th quarter
|
All (n = 152)
|
|
105.4 s
|
85.3 s
|
46.2 s
|
45.2 s
|
70.1 s
|
|
10.1 mGy
|
8.2 mGy
|
4.4 mGy
|
4.4 mGy
|
6.8 mGy
|
The difference in radiation exposure time between the first and the third quarters
was statistically significant (p = 0.0073) as was for the first and last quarters (p = 0. 0082).
The operative blood loss was between 50 and 200 mL (average: 110 mL). There were no
intraoperative complications or conversions to open surgery. The LOS ranged from 2
to 7 days (average: 6.0 days), and the mean time until leaving the bed was 1.8 days
([Table 4]). The difference in blood loss and LOS for different quarters was not statistically
significant.
Table 4
Blood loss and length of hospital stay (LOS) for each quarter and for the whole series
|
1st quarter
|
2nd quarter
|
3rd quarter
|
4th quarter
|
All (n = 152)
|
|
Average blood loss (mL)
|
113.3
|
115.0
|
106.6
|
107.1
|
110.0
|
|
Average LOS (d)
|
5.9 d
|
6.3 d
|
5.4 d
|
6.5 d
|
6.02 d
|
The VAS score was reduced from 7.9 points preoperatively to 2.1 points at 1 year postoperatively.
The ODI score improved from a preoperative severe disability (mean: 52.1%) to moderate
disability (25.2%) at 1 month up to a minimum disability (mean 17.9%) at 1 year postoperatively.
The difference in the functional outcome for the different quarters was not statistically
significant.
In two cases, there was a dural tear intraoperatively. No perioperative infections
were observed. In four patients, there was malposition of a screw, but only one was
symptomatic. Cage subsidence was observed in three cases. In three patients with severe
osteoporosis, loss of correction and screw failure were observed during the follow-up.
Discussion
In surgery, a learning curve is defined as the time taken and/or the number of cases
required by a surgeon to obtain good results and to become proficient, for example,
to reduce operative time, to reduce estimated blood loss, and to reduce the morbidity,
adverse events, and complication rate. The literature is not conclusive whether the
learning curve of MISS should be defined as steep or shallow as both are used with
opposite meanings.[8]
[9] The term steep learning curve is often used in informal language to mean a difficult,
challenging learning process. However, most sources and our own interpretation define
a learning curve as a plot showing proficiency as a function of the number of repetitions,
so a steep increase would mean that quick increment of the necessary skills and performance
are achieved in fewer repetitions[9]
[10] ([Fig. 1], curve A). Consecutively, good results are obtained for a shorter period of time—with
a smaller number of cases. Shallow learning curve, on the other hand, would require
more time and more cases to reach good results and performance ([Fig. 1], curve B). Regarding the learning curve for MISS, we believe it is multifactorial
and should not be defined simply as steep or shallow. It would be more useful and
informative if we analyze the learning curve for different parameters of the surgical
procedure separately.
Fig. 1 Learning curve. Curve A: steep learning curve. Good results (performance) are achieved
quickly, for a short period of time (small number of cases). Curve B: shallow learning
curve. Good results (performance) are achieved for a longer period of time (more cases
are required to obtain good results).
As discussed in the literature, the advantages of MISS are less muscle and tissue
disruption, minimized blood loss, reduced rate of surgical site infection, reduced
recovery period, shortened hospital stays, and higher patient satisfaction.[1]
[2]
[3]
[4]
This is not surprising given that MISS employs techniques such as tubular retraction
or minimal skin incisions, which preserve the contralateral ligament and bony attachments
of paraspinal muscles, thereby reducing potential bleeding. The minimal muscle dissection
and bone removal also will reduce complications attributed to blood clot accumulation
and tissue fluid accumulation.[5] Several studies demonstrate statistically significant or highly significant reductions
in intra-/perioperative blood loss in the MI-TLIF cohorts compared with their open
TLIF groups ranging from mean values of 100 to 456 mL in the MI-TLIF cohorts versus
450 to 961 mL in the open TLIF group.[6]
[7] Villavicencio et al[2] reported lower estimated blood loss (163 mL) and shorter hospitalization (3 days)
in MI-TLIF compared with open TLIF (366.8 mL and 4.2 days). In our group, the estimated
blood loss was 50 to 200 mL, which is comparable with those reported in the literature
and supports the finding that minimally invasive techniques significantly reduce tissue
damage and intraoperative blood loss. Decreased exposure surface and limited muscle
disruption also significantly reduce the opportunity for bacteria entry and hence
surgical site infection.[3] In our series, the average LOS was 6.02 days. Patients were verticalized after an
average of 1.8 days. No operative infection was observed. Other studies also show
a significantly shorter time to ambulation in the MISS groups compared with the open
technique groups.[11] Eckman et al[12] discharged 73% of their 1,114 patients on the day of the MI-TLIF, meaning 808 of
their patients were mobilized on the day of the surgery. In seven studies, LOS was
significantly shorter in the MI-TLIF cohorts with a mean LOS of 3 to 9.3 days compared
with the open TLIF groups with a mean LOS of 4.2 to 12.5 days.[4]
[7]
Minimized surgical trauma and reduced paraspinal muscle dissection are likely responsible
for the significant reduction in postoperative VAS and ODI scores in the MISS cohorts
versus the open technique cohort. In a meta-analysis by Vazan et al,[1] 11 studies reported VAS on follow-up of 1.0 to 3.4 in the MI-TLIF group and 1.2
to 7.5 in the open TLIF group. A meta-analysis by Phan et al reported a mean difference
in VAS back pain scores of 0.4 points lower for MI-TLIF, and an ODI score 2.2 points
lower for MI-TLIF compared with open TLIF.[5] In our group, the mean preoperative VAS score was 7.9 and the average postoperative
score was 2.1 at the last follow-up. The ODI score improved from the average preoperative
of 52.1% to 17.9% at the last follow-up.
Contemplating on the learning curve of MISS, we analyzed how the above-mentioned parameters
of the surgical procedure changed with more surgical experience. In our series, there
was no significant difference in the average operative blood loss and mean LOS between
the four consecutive quarters ([Table 4]). There was also no significant difference in functional outcome—the mean postoperative
VAS score was 0.2 lower and the mean postoperative ODI score was 1.1 lower for the
later 76 cases. That is why we believe that the benefits discussed not only are indisputable
and evidence based but also can be achieved with the very first cases without significantly
depending on the surgeon's experience with the MISS technique. Therefore, the learning
curve of MISS for these parameters of the surgical procedure should be defined as
steep because favorable results are achieved at the very beginning and after only
a few repetitions.
However, there is still controversy regarding the possible disadvantages of MISS—longer
operating times, higher intraoperative radiation, and a potentially higher risk of
cage and pedicle screw misplacements. There are studies that show significantly shorter
operation times in the open surgery groups (90–250 minutes) compared with the minimally
invasive group (135–375 minutes).[1] No difference in length of surgery is shown by Schizas et al,[6] Brodano et al,[13] and other studies. Phan et al,[5] in a systematic review of the literature, also found no significant difference in
operation time between the MI-TLIF and open TLIF cohorts (median duration of 185 minutes
for minimally invasive compared with 186 minutes for the open procedures). Chang et
al also found no significant difference in ODI and operation time between MIS and
conventional open surgery.[14] Wang et al,[15] however, found a significantly shorter operating time for MI-TLIF (127 ± 25 minutes)
compared with open TLIF (168 ± 37 minutes) in obese patients. Concerning radiation
exposure, most of the studies have shown significantly longer radiation exposure times
during MISS surgery (range: 45.3–106 seconds) compared to conventional open procedures
(range: 24–39 seconds).[2]
[15]
[16] Phan et al, in their literature review and meta-analysis, also reported that the
X-ray exposure time was significantly higher in the MI-TLIF group compared with the
open TLIF group by 37 seconds.[5]
In our opinion, these results are due to the relatively shallow learning curve associated
with MISS techniques for these parameters of the surgical procedure. In our group,
the mean operative time was 141.6 minutes, which is comparable with results from other
literature reports. However, we found a decrease in operative time with the increasing
surgical experience of our team—mean duration of surgery for the first quarter of
38 cases was 155 minutes, 143.21 minutes for the second quarter, 134.4 and 133.7 minutes,
respectively, for the third and last quarters ([Fig. 2]). This means that we needed ∼76 cases to achieve optimal reduction of our time for
surgery. The operative time for the third and fourth quarters was significantly shorter
than that for the first 38 cases.
Fig. 2 Operative time in minutes for the 1st, 2nd, 3rd, and 4th quarter of patients.
Regarding the X-ray exposure in our series, the average time of radiation exposure
was 70.05 seconds or 6.84 mGy ([Table 3]), which seems to be significantly different from that of the open surgery techniques.
However, the exposure time also showed a significant decrease with the increasing
surgical proficiency—average 105.4 seconds or 10.06mGy for the first quarter of patients,
85.25 seconds or 8.26 mGy for the second, 46.2 seconds or 4.44 mGy and 45.18 seconds
or 4.35mGy for the third and last quarters ([Fig. 3]). We have achieved 50% decrease of radiation exposure, with the results of the latter
two quarters comparable to those of conventional surgery.
Fig. 3 Time of X-ray exposure in seconds for the 1st, 2nd, 3rd, and 4th quarter of patients.
Comparing the results of the four consecutive groups, it is evident that there was
an important decrease in the duration of surgery and radiation exposure between the
first three quarters and almost no difference between the third and fourth group of
patients. There was no significant difference in functional outcome regarding the
VAS and ODI scores when comparing the four timeline groups. On the other hand, most
of the complications were observed in the first two quarters—4 cases of screw malposition,
2 dural tears, 2 cases with loss of correction, and 2 cases with cage subsidence.
One case with loss of correction and 1 case with cage subsidence were observed in
the third quarter and no complications were observed for the last 38 patients.
Therefore, we can conclude that the learning curve for some parameters of the surgical
procedure such as duration of surgery, intraoperative radiation, and a potentially
higher risk of cage and pedicle screw misplacements should be defined as shallow,
and acquisition of necessary skills, experience, and proficiency took ∼76 cases in
our series.
The improved efficiency could be explained with the increasing experience of the surgeon
and the surgical team, which led to familiarity with the operative steps, improvement
in the workflow ergonomics, and evolution of the operative techniques. This is evident
if we analyze the reduction of operative time and radiation exposure with gaining
more surgical experience. For example, for the first quarter of patients, an average
of 67 fluoroscopic shots per surgery were used compared with an average of 29.6 fluoroscopic
shots per surgery for the third quarter, and 29 fluoroscopic shots per surgery for
the fourth quarter. We can conclude that increasing the proficiency of the team over
time led to more confidence of the surgeon, less need for fluoroscopic control, and
consequently reduction of the operative time. This improvement of efficiency cannot
be attributed only to the ameliorated skills and confidence of the surgeon. The better
results came with the improvement of workflow and implementation of new operative
techniques. For example, the use of two parallel to the midline covering two to three
pedicle skin incisions instead of multiple small incisions for each pedicle ([Fig. 4]) led to much better tactile sensation and anatomical orientation for the surgeon
with a resulting significantly shorter operative and radiation exposure times.
Fig. 4 Two-level transforaminal lumbar interbody fusion (TLIF) with two parallel to the
midline skin incisions instead of multiple small incisions.
Another example is the introduction of the “four-hand surgical technique.” The surgical
steps are not performed by one surgeon (Senior Surgeon) with the help of an assistant,
but both surgeons work simultaneously and sequentially on the same pedicle screw sharing
the different steps to save time. For example, surgeon 1 protects the k-wire and the
soft-tissue retractors, and the second surgeon is tapping; then the second surgeon
inserts the screw while the first one takes care of the k-wire. The “four-hand surgical
technique,” trained scrub nurse, and well-designed operative plan and ergonomics allowed
a continuous workflow with significant reduction of time per screw and radiation exposure.
Fifty-six cases of the latter two quarters were performed in this manner and this
is probably one of the main reasons for the significant decrease of time of surgery
and radiation exposure in comparison with the first two groups.
In our opinion, one of the most important factors for better results was the accurate
patient selection. When analyzing our series, we observed that more of the cases were
performed at the beginning of the time period—91 cases for the first half (from April
2013 to December 2015) and 61 cases for the second 3 years (from January 2016 to December
2018). In the second half of the study, better patient selection led to better results
regarding complications, shorter time of ambulation, and shorter duration of surgery
and radiation exposures. In selected pathologies, time of surgery and radiation exposure
time were even shorter than the reported times for open techniques.
This conclusion is supported by other reports. For example, Schizas et al[6] found an average decrease of operative time by 1.8 hours (from 6.1 to 4.3 hours)
when comparing the first and last third in their MI-TLIF series. In a study by Lee
et al, the mean operative time in the early group was 187.2 minutes, decreasing significantly
to 132.3 minutes in the later groups. Similarly, the mean fluoroscopy duration was
74.4 seconds in the early group, which decreased to a mean of 29.9 second in the later
groups.[17]
Regarding our experience, we have to also emphasize that MISS might be particularly
beneficial in obese patients with a BMI > 30. In our group, MISS was used in 24 cases
of obese patients. The mean operative time in these cases was 146.6 minutes compared
with 142.8 minutes for the whole series. The average blood loss was 140 mL and the
mean radiation exposure time was 73.2 seconds. Other studies also found less blood
loss and lower complication rates, shorter operating time and hospital stay, and reduced
local pain after MI-TLIF in obese patients.[1]
[15]
Conclusion
Minimally invasive percutaneous spinal fixation techniques have some clear advantages,
such as reduction of the iatrogenic intraoperative tissue trauma, thus minimizing
blood loss and postoperative pain, and shortening the LOS. We believe that these soft
tissue–related benefits can be observed in the initial cases and are not related significantly
with the surgeon's experience with the MISS technique.
With the acquisition of more experience, some disadvantages of MISS techniques such
as longer operative time and longer X-ray exposure can be substantially and significantly
reduced and become comparable to open techniques.
Surgical experience, familiarity of the team with the MISS instrumentation, and good
patient selection are crucial for achieving all the benefits of MISS techniques and
make them a reasonable and less invasive alternative to conventional open surgery
techniques.
