AMPUTATION IN FRACTURES OF THE TIBIA

J. H. HICKS, BIRMINGHAM, ENGLAND

From the Birmingham Accident Hospital

Fractures of the shaft of the tibia in adults are notoriously difficult to treat and are
sometimes dangerous. Poor function, non-union, skin lossand infection are the great hazards,
and an awkward dilemma is posed by the fact that the treatment least likely to give one of
these bad results is often the most likely to give another. Amputation, both primary and
secondary, should be included among the bad results of any series since these represent
complete failures. It is not uncommon to find secondary amputation referred to as though
it were usually a complication of operative treatment but facts and figures are difficult to
obtain. Dehne, Deffer, Hall, Brown and Johnson (1961), in a series of over 200 cases treated
conservatively, had no secondary amputations. On the other hand, Holderman (1959), in a
series of forty-five cases treated mostly conservatively, had three secondary amputations
(6 per cent) in addition to a late primary amputation due to a tight plaster. Scott (1962),
also in a predominantly conservative series, recorded two secondary amputations in 236
cases (08 per cent). The survey of Holdsworths 99 per cent conservative series by Ellis
( 1956), which in other respects is comprehensive, only tabulates late primary amputations
(23 per cent), but two secondary amputations (05 per cent) are mentioned in footnotes and
it is by no means certain that these were all. Other authors only deal with fractures at the easy
end of the range. For example, Alms (1962) includes in his series for intramedullary nailing
mostly the non-comminuted and the mildest compound fractures and does not concern
himself with results among the cases he rejects. Yet as Ellis (1958) pointed out the easy
fractures can be expected to give good results. The bad fractures are the ones that tend to
do badly.
The fact that the tragedy of amputation can occur with conservative treatment-even of
simple fractures-is evident not only from Holdermans series but also from two of the cases
described below. The suspicion remains, however, that operative treatment of the original
fracture is likely to raise the incidence of complications. This study, therefore, of all the
amputations done at the Birmingham Accident Hospital over a period of twelve years is of
interest, since the policy of the Hospital is well known to be in favour of the much criticised
internal fixation for the more severe cases of fractured tibia.

MATERIAL

The number of fractured shafts of tibia in adults treated in twelve years was over 1,400.
The number of legs amputated, after excluding amputations entirely due to some concomitant
injury or its complications, was fifty-five. The word adult refers to persons of fifteen or over.
By shaft is meant all cases except those in which the centre of the fracture was less than
four and a half inches from the knee joint or three inches from the ankle joint. Every degree
of severity of injury up to and including traumatic amputation had to be counted since
no clear line of demarcation could be made between the almost severed limb and the gross
compound fractures of the tibia with absent distal pulses, some of which have been treated
with success.

RESULTS
The incidence of amputations is given year by year in Table I. Division into secondary
and primary amputations is the first step towards separating those cases in which the fracture
treatment could possibly have been responsible for the need to amputate from those in which

388 THE JOURNAL OF BONE AND JOINT SURGERY

 AMPUTATION IN FRACTURES OF THE TIBIA 389

the deciding factor was entirely the severity of the injury. Even this division is not clear-cut
because of an intermediate group of eight late primary amputations which, in the present
paper, has been included with primary amputations. In six of these the circulation was absent
on admission; nevertheless an attempt had been made to preserve the limb. In one case the

TABLE I
INCIDENCE OF AMPUTATION DUE TO THE FRACTURE, OR lTs
COMPLICATIONS, OR ITS TREATMENT

Secondary amputations Primary amputations

Year ----- - - ------ - ---

Numbers Per cent Numbers Per cent

1951 3

1952 2 7

1953 3 - 23 4 35

1954 21 2

1955 4J 5J

1958 I 3

1959 2. 0.7 5 17

1960 0

1961 I I

1962 0J 0J

TABLE II
CAUSES FOR SECONDARY AMPUTATIONS AND THEIR RELATIONSHIP TO TREATMENT

Number of cases
Skin necrosis, sepsis, no union; with or without failed
skin grafting . . . . . . . . 9

Skin necrosis threatening the above sequence . .

Pure sepsis . . . . . . . .

Gas gangrene. . . . . . . .

Septic bone graft after initial healing . .

Mechanical breakdown . . . . . .

Refracture: one with recurrence of sepsis . . . 2

Large bone defect . . . . . . .

Late necrosis or fibrosis of deep tissues . . . 3

circulation disappeared during or after the primary treatment. The other case was of old
osteomyelitis with fracture.
In Table lithe immediate causes for the twenty secondary amputations are analysed to
show their relationship to treatment.

VOL. 46 B, NO. 3, AUGUST 1964

390 J. H. HICKS

DISCUSSION
Table I shows that the secondary amputation rate (at least in recent years) is of the same
order as in conservatively treated series (Ellis 1956, Holderman 1959, Scott 1962).
Table II confirms that skin necrosis, shown in Figures 1 and 2, is justifiably feared as
the commonest starting point of a chain of events that leads ultimately to amputation. Primary
skin loss at the time ofthe accident was never a factor in this series-an observation at variance
with current teaching. Sepsis was rare as a primary complication ; nearly always it gained
access afterwards ; and it was a factor influencing the decision to amputate in two-thirds of
the cases. In many cases the reason for amputation was that, with non-union present or
incipient, infection precluded any attempt at bone grafting. Corrosion of metal was not a
cause of amputation. Osteoporosis, contractures or a painful atrophic limb, sometimes quoted
as causes for amputation, accounted for three in this series. Sixteen of these twenty fractures
had been compound, most of them severely, and there is, in consequence, ample to account
for the amputations without necessarily incriminating the treatment. Conservative treatment
is not above suspicion, at least in our hands ; one of the simple fractures had been treated
conservatively yet nevertheless developed skin necrosis and secondary sepsis. One hopeful
feature of the series is that in recent years (with the possible exception of the last of the causes
listed) cases corresponding to every group in Table II have been successfully treated and
permanent healing and sound union obtained. Two such fractures are to be seen in Figures
1 and 3. It may be this improvement in treatment that has been responsible for the lowering
of our amputation rate revealed in Table I.
Reverting to Table I, it will be observed that in 1956 the incidence ofsecondary amputation
dropped from twenty-three per thousand to seven per thousand. This could be accounted for
by a reduction in the sepsis rate or by a general decrease in the severity of fractures arriving
at the hospital or by a greater readiness to do primary amputations, but there are reasons to
believe that none of these explanations is the essential one. Thus Figure 3 shows that we are
certainly not escaping the bad fractures nor are we making the severity of a fracture an excuse
for amputation, and Table I confirms that our primary amputation rate is diminishing
concurrently. The only remaining explanation is that our treatment for the complications
has become more effective.
One difference between the early and the late periods of this survey was that our attitude
to internal fixation had progressively hardened. Not only was the proportion of plated
fractures higher in the later period but the rule that fixation should be rigid or not done at all
was more constantly obeyed, and about the time of the transition between these two periods
the largest plate ever to be used on the tibia was introduced. Initial treatment ofthe compound
fracture in Figure 3, for instance, was fixation by two plates. This must not be interpreted
as a recommendation for indiscriminate internal fixation. Our methods of handling both the
original fractures and the complications differ from the orthodox in so many details that,
until they are well understood and completely accepted, internal fixation must still be
discouraged elsewhere. An earlier analysis of some of the same material, for instance, has
shown that the incidence of sepsis is definitely increased by plating (Hicks 1957). Our
diminishing number of amputations, however, indicates that internal fixation properly hand/ed,
far from interfering the treatment
with of sepsis, can be made to facilitate it.
The most common criticism of internal fixation is that, because of sepsis, the risk of
amputation is increased. In the present series, however, amputation as a result of sepsis was
done in only 1 per cent (l8 per cent up to 1956 and 04 per cent since). The assumption
that all of these were the result of treatment is unjustified, just as the assumption that all
amputations in a conservative not due to the treatment.
school are There is some hint that
conservative regimes have their own positive hazards. Thus the atrophic, painful limb is
often quoted as one of the causes of late amputation whereas this factor was rare in the
present series. Furthermore, the comparatively high late primary amputation rate due to

THE JOURNAL OF BONE AND JOINT SURGERY

 AMPUTATION IN FRACTURES OF THE TIBIA 391

Li
FIG. 1 FIG. 2
Figure l-Ischaemic death of skin caused partly by stripping of the skin from
its deep attachments and partly by the surgical incision. This case was brought
to a successful conclusion. Figure 2-Skin defect after sloughing of necrotic
skin, secondary sepsis and failedskin graft. Dead bone is present and there is
no union. This leg was amputated.

r. I.i
\ f,.i
1

I,,

I
FIG. 3
The clinical appearances and radiographic tracings of a fractured tibia from the recent half of the
survey. Although the foot was described as probably not viable this leg was preserved. Every
complication (skin necrosis, sepsis, large bone defect and delayed union) was encountered and all were
overcome. Final healing of the wounds and consolidation of the fracture were achieved after one year
and the patient was back at work after fourteen months. The initial treatment was by internal fixation.

VOL. 46 B, NO. 3, AUGUST 1964

392 J. H. HICKS

circulatory failure in Elliss series, treated mostly in plaster, suggests another hazard and the
means by which it is brought about. While the subdivision into inevitable amputations
and iatrogenic amputations remains impossible to assess, the above overall percentage
must remain as the basis for comparison.

SUMMARY
I. The conservative school of treatment of fractures of the tibia, which bases part of its
criticism of internal fixation on the ultimate risk of amputation, does not often publish its own
rates of amputation.
2. Statistics from a hospital that treats one-third of closed fractured tibiae and two-thirds of
compound fractures by internal fixation are therefore put up as a basis for criticism.
3. Comparisons are made with the few available statistics in the literature of conservative
treatment.
4. Almost all of the causes for secondary amputation are now curable and in recent years
the number of limbs being saved is increasing.

REFERENCES

ALMS, M. (1962): Medullary Nailing for Fractures of the Shaft of the Tibia. Journal of Bone and Joint Surgery,
44-B, 328.
DEFINE, E., DEFFER, P. A., HALL, R. M., BROWN, P.W., and JOHNSON, E. V. (1961): The Natural History of
the Fractured Tibia. Surgical Clinics of North America, 41, 1,495.
ELLIS, H. (1956): A Study of Some Factors Affecting Prognosis Following Tibial Shaft Fractures. Oxford: Thesis,
Bodleian Library.
ELLIS, H. (1958): The Speed of Healing after Fracture of the Tibial Shaft. Journal of Bone and Joint Surgery,
40-B, 42.
HICKS, J. H. (1957): The Relationship between Metal and Infection. Proceedings of the Royal Society of Medicine
(Section of Orthopaedics), 50, 842.
HOLDERMAN, W. D. (1959): Results Following Conservative Treatment of Fractures of the Tibial Shaft. American
Journal of Surgery, 98, 593.
SCorr, J. C. (1962): In Modern Trends in Orthopaedics No. 3. Edited by J. M. P. Clark. London: Butterworths.

THE JOURNAL OF BONE AND JOINT SURGERY